'A-600/5-75-004
MARCH 1975
                           Socioeconomic Environmental Studies Series
     Analysis of Nonpoint-Source  Pollutants
            In The  Missouri Basin Region
I
55
%
                                                 UJ
                                                 CD
                                  Office of Research and Development
                                  U.S. Environmental Protection Agency
                                  Washington, D.C. 20460

-------
                RESEARCH  REPORTING SERIES
 Research  reports  of the  Office  of Research  and Development,
 Environmental  Protection Agency,  have been  cirouped into  five
 series.   These five broad categories were established to
 facilitate  further  development  and application of environmental
 technology.  Elimination of  traditional grouping was consciously
 planned to  foster technology transfer and a maximum interface
 in related  fields.   The  five series are:

      1.   Environmental Health Effects Research
      2.   Environmental Protection Technology
      3.   Ecological Research
      4.   Environmental Monitoring
This report has been assigned to the SOCIOECONOMIC ENVIRONMENTAL STUDIES
series.  This series describes research on the socioeconomic impact of
environmental problems.  This covers recycling and other recovery
operations with emphasis on monetary incentives.  The non-scientific
realms of legal systems, cultural values, and business systems are
also involved.  Because of their interdisciplinary scope, system
evaluations and environmental management reports are included in this
series.

This report has been reviewed by the Office of Research and
Development.  Approval does not signify that the contents
necessarily reflect the views and policies oftthe Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
Document is available to the public through the National Technical
Information Service, Springfield, Virginia  22151.

-------
                                           EPA-600/5-75-001*
                                           March 1975
ANALYSIS OF NONPOINT-SOURCE POLLUTANTS IN THE
               MISSOURI BASIN REGION
                          By

                  Dr.  A. D. McElroy
                    Dr.  F. Y. Chiu
                     Dr.  A. Aleti
                 Contract No. 68-01-1863
              Program Element No. 1BA030
                     ROAP 16AFN03
                      Project Officer
                   Dr. Marshall Rose
        Washington Environmental  Research Center
                 Washington, D.C.  20460
                       Prepared for
           Office of Research and Development
          U.  S. Environmental Protection Agency
                Washington, D.C.  20460

-------
                             ABSTRACT

    A study was conducted of nonurban, nonpoint sources of pollution in
the Missouri Basin utilizing a unique, computer-based data system.  The
Data Bank contains extensive information on land use, soil, climate,
water quality,  and other parameters in the Missouri Basin.  The current
study was concerned with the adequacy of the Data Bank relative to develop-
ment of nonpoint pollution models.

    Evaluation of information in the Data Bank yielded detailed land use
and water quality profiles in the basin.  The results show that the Data
Bank is a useful base to depict the  basinwide  relations between various
land uses and water quality.  Regression equations of these relationships
were developed for individual subbasins as well as for the entire basin.
The regression equations, and the  accompanying statistical analysis  of
significance, indicated that the water quality  parameters NO3, BOD, and
turbidity correlate well with land use on a basinwide basis.  Other
parameters, namely phosphorus, dissolved oxygen, and dissolved solids,
did not correlate well with land use.

    This report was submitted in fulfillment of Project Number 16AFN03,
Contract Number 68-01-1863, by the Midwest Research Institute, under
the sponsorship of the Environmental Protection Agency.  Work was
completed as of January 1975.
                                 11

-------
                                 CONTENTS
Abstract	     it

 List of Figures	    iii

 List of Tables	   viii

Acknowledgement  ..... 	  .....  	     ix

 Introduction	      1

 Objectives and Scope of Phase II(A) Study 	      3

   General  Objective 	      3
   Specific Objectives 	  .  	      3
   Scope	      4

 Background Discussion of Phase I Study  	  ...      5

 Approach for  Phase II(A) Study  	      7

   Rationale . 	 ..................      7
   Method of Study  	      8
   Tasks	      9

 Results	     15

   Basinwide Land Use Characterization 	  .....     15
   Basinwide Water  Quality Characterization  ......  	     19
   Stream Water Quality Characteristics  	     30
   Regression  Analysis ..... 	     57

 Conclusions and  Recommendations 	 ......     76

   Conclusions	     77
   Overall  Assessment	     81
   Recommendations  	 .........  	     82
                                 iii

-------
                           CONTENTS (Concluded)




                                                                    Page





Appendix A - Data Sources	   89





Appendix B - Data Organization	   96




Appendix C - Data Processing Techniques	  103




Appendix D - Computer Programs	106
                                   iv

-------
                                  FIGURES

rigure                                                               Page

   1      Protocol of Statistical  Evaluation of  Data  	     10

   2      Distribution of Cattle in Missouri Basin 	     20

   3      Distribution of Hogs in  Missouri Basin	     21

   4      Distribution of Cattle Per  Square Mile in Missouri Basin,
            Station Mean Values	     22

   5      Distribution of Cattle Per  Square Mile in Missouri
            Basin	     23

   6      Distribution of Hogs Per Square  Mile in Missouri Basin,
            Station Mean Values	     24

   7      Fertilizer Use (N and PC^)  in Missouri Basin	     25

   8      Herbicide Use (tons) in Missouri Basin ... 	     26

   9      Insecticide (tons) Use in Missouri Basin 	 ...     27

  10      Fertilizer (N and P) Use in Missouri Basin (tons/
            miles2), Station Mean Values 	     28

  11      Dissolved Oxygen (mg/liter) in Streams of Missouri
            Basin	     33

  12      Nitrate (mg/liter) in Streams of Missouri Basin  ....     34

  13      BOD (mg/liter) in Streams o£ Missouri Basin	     35

  14      Total Colifotm (No./lOO ml) in S* - ,:uis of Missouri
            Basin	.	     36

-------
                            FIGURES (Continued)




Figure                                                              Page




  15
16
17

18
19
20
21
22
23
24
25
26
27
28
29
30

31

32

Distribution of Dissolved Solids in Missouri Basin . . .
Distribution of BOD (Ib/day) Per Square Mile in

Water Quality Along the Main Stem of Missouri River . . .








Water Quality Along Platte River 	

Regression of BOD-MGL Versus Cattle Per Square Mile . . .
Regression of BOD-PPD Per Square Mile Versus Livestock

Regression of BOD-PPD Per Square Mile 'Versus Cropland

Regression of NOo-PPD Per Square Mile Versus Cattle

38

39
40
41
42
43
44
45
46
47
48
49
50
66

67

68

69
                                    vi

-------
                            FIGURES  (Concluded)

Figure

  33      Regression of Turbidity, JU Versus Cattle Per Square
            Mile	    70

  34      Regression of NC^-PPD  Per  Square Mile Versus Nitrogen
            Per Square Mile	    71

  C-l     Data Processing  Sequence	   105
                                  vii

-------
                                  TABLES

Table

  1       Summary of Land Use Data by Subbasins for 1969  .....     16

  2       Annual Land Use Loading in Missouri Basin, Station  •
            Mean Values	     18

  3       Water Quality in Missouri Basin  	     29

  4       Pollutant Burden in Missouri Basin (Ib/mile^/day)   ...     31

  5       Summary of Water Quality and Pollutant Burden in
            Missouri River Basin 	     32

  6       Pollutant Potential of Various Farm Animals  	     59

  7       Pollutant Loading Versus Land Use:  Linear Regression
            Analysis	     60

  8       Water Quality Versus Land Use:  Linear Regression
            Analysis	     61

  9       Pollutant Loading Versus Land Use:  Logarithmic Regres-
            sion Analysis	     62

 10       Water Quality Versus Land Use:  Logarithmic Regression
            Analysis	     63

 11       Summary of Multiple Regression Analyses	     74

 B-l      County File Information Sheet, 1969	     98

 B-2      Station File Information Sheet, 1969 	  .....    100
                                  viii

-------
                  ACKNOWLEDGEMENT

    The study  described in this document was performed under
contract to  the  Environmental Protection Agency,  by Midwest
Research Institute in the Physical  Sciences  Division,  Dr- H.
M.  Hubbard, Director.   The technical direction of the program
was provided by  Dr. A.  E.  Vandergrift,  Assistant  Director,
Physical Sciences Division.  Dr. A.  D. McElroy, Head, Treat-
ment and Process Control Section, served as program  manager.
Dr. Adi Aleti, Dr. Shen-Yann Chiu,  and Dr. A. D. McElroy
are the principal  authors of the report.
                                  ix

-------
                               SECTION I

                              INTRODUCTION

     Water pollution from nonpoint sources is an area of justified concern
to the Environmental Protection Agency and other water resources  planning
agencies in the nation. This concern is especially significant  in the
Missouri River Basin where the primary land use is agriculture. Methods
to assess the degree of water quality degradation from nonpoint  sources
are currently less than adequate.  Consequently, planning agencies are  in
urgent need of predictive tools to effectively implement the necessary
water quality management strategies in various river basins.

     Midwest Research Institute recognized this need several years ago,
and initiated Phase I of the current program through the sponsorship of
the Environmental Protection Agency (EPA Contract No. 68-01-0159). This
Phase I study resulted in the development of a unique data bank for the
Missouri River Basin comprising land use data, soil, climate, and other
physiographic data, and water quality data in the basin. Land use data,
soil data, and climatic and other physiographic data including crops,
fertilizer use, and livestock data were based on each county (County Mas-
ter File). These data were transformed to watersheds represented by
selected water quality monitoring stations (Station Master File)  which
also contained water quality data. This data base was organized in a
computer processible form. Because of the potential utility of the data
base in characterizing the interrelationships between parameters of land
use and water quality in the basin, the current Phase II(A) study was
initiated under EPA Contract No. 68-01-1863. This study, in a relatively
modest effort, was designed to further characterize data in terms of
quality and quantity, and for potential utility in existing or new water
quality models. Thus, the Phase II(A) program was set up to determine,
the nature and merit of the data base for more comprehensive and de-
tailed analysis.

-------
     This report presents and diuexists the results of Phase II(A)  study
in the following sections:

     II.  Objectives and Scope of Study

     III.  Background Discussion on Phase I Study

     IV.  Approach to Phase II(A) Study

     V.  Results

     VI.  Conclusions and Recommendations

-------
                               SFCTION II

               OBJECTIVES AND SCOPE OF PHASE II(A)  STUDY

     The following are the general and specific objectives and  scope of
Phase II(A) study.

GENERAL OBJECTIVE

     The general objective of Phase II(A) was preliminary development and
testing of appropriate water quality models for the Missouri River Basin,
with the available data generated during Phase I. This  broad objective
was redefined early in Phase II(A), and a more limited  objective was set:
To thoroughly analyze the data and test its validity  in terms of simple
conceptual models, and thus to establish a firm basis for either the de-
velopment (in anticipated continuing study) of comprehensive models, or
for extension of the data base as required to improve its utility. Speci-
fic objectives, consistent with the general objective,  are presented
below.

SPECIFIC OBJECTIVES

     The specific objectives of Phase II(A) were:
     • To evaluate the adequacy of the data base of Phase I as  inputs to
       models.

     • To conceptualize and specify the type and form of these  models.

     • To make preliminary tests of the feasiblity  of models with the
       available data.

     • To appraise the tradeoffs between developing models which work with
       available data versus extended collection of additional  data.

-------
SCOPE

     The data base contains extensive information on the agricultural prac-
tices which are the predominant sources of nonpoint pollution in the Missouri
River Basin. In addition, essential information on climate, terrain, soil,
water quality, and other pertinent factors is contained in the data base.
The  study  involved analyses of these multiple factors for the entire Missouri
River Basin as well as  for selected subdivisions.

     Thirteen subbasins were  characterized according to land use and water
quality, and the water  quality/land use patterns along significant river
-basin segments were examined. The characterizations involved both computer
assisted analysis and correlation of data, and collection and plotting
of the various kinds of data  which can be readily retrieved from the Data
Bank.

     The data base was  thoroughly analyzed in terms of its comprehensive-
ness and quality of data for  developing conceptual models, with emphasis
on the impacts of land  use practices on water quality in the basin.

-------
                              SECTION III

                 BACKGROUND DISCUSSION OF PHASE I  STUDY

     In a preliminary effort to define and evaluate nonurban, nonpoint
source pollutants in the Missouri River Basin,  MRI compiled  large  quanti-
ties of data for the region during the Phase I  study.  These  data were ob-
tained from various data sources, and are discussed briefly  in Appendix A.

     Because the majority of land use and economic data are  on a county
base, a computerized data storage system was developed which relates county
segments to watersheds.  This system permits determination  of correlations
between land use data and water quality data, assessment of  the contribu-
tion of agriculture and other nonurban pollutant sources to  the degrada-
tion of streams and rivers in the Missouri River Basin, and  will eventually
be valuable for predicting the improvements in water  quality that  would re-
sult from specific nonurban control practices.

     The data compiled in Phase I contain substantial amounts of pertinent
information, a large portion of which is in computer-processible form.
Data categories available include:  water quality, hydrology, climatology,
land use, topography, soil classification, livestock,  pesticide use, ferti-
lizer use, and other important economic data.

     The organization of the data base is briefly discussed  in Appendix B.

     In order to interpret the significance of these  data  for the  study
area, the Phase II(A) study was initiated. The data  interpretation involved
a consideration of the nature of the basin characteristics—both intensive
and extensive—in generating and transporting pollutants.  The Missouri
River Basin is predominantly agricultural in nature—agriculture occupies
some 390,000 miles2 (250 million acres or 100 million hectares) of farm
land. This area represents about 85% of the total land area  in the basin.
Several kinds of pollutants are identified as significant  in basin water
quality. These include nutrients, pesticides, salts,  sediments, organic
residues, biodegradable wastes, and microbial pathogens.

-------
     The quantity of pollutants generated from nonpoint sources depends
on the land use. For example, row crop production typically yields rela-
tively large quantities of eroded sediments, while hay and forage crop
production is less susceptible to water erosion. Similarly, feedlots  in
the vicinity of streams are expected to contribute heavily to stream  pol-
lution in terms of nitrates, biodegradable organics, and bacteria.

     Sensible planning for control of pollutant emissions from agricul-
ture and of water quality in surface and groundwater streams and reser-
voirs requires quantitative knowledge of pollutant loads as functions of
several factors, and the inventory of supportable quantitative facts  and
data for assessing the fate of pollutants in the agricultural environment
must be greatly strengthened. Prediction models for nonpoint pollution
require a thorough knowledge of not only the land use characteristics and
basic land characteristics, but also of pollutant generation and trans-
port mechanisms. The validation of these models requires, in addition,
data on water quality itself as well as on the dynamic behavior of the
stream. In the past, discrete models have been developed on the various
aspects of pollutant generation and transport, but these models have  not
been synthesized into a single useful form to comprehensively evaluate
nonpoint pollution.

     Thus, the functions generated in the current Phase II(A) study were
dictated to a large extent by the available data and by model requirements
for planning purposes. Consequently, the data base was analyzed in detail
for quality and for the extent of its usefulness in models, and to deter-
mine if additional data should be acquired and incorporated in the data
base.

-------
                               SECTION IV
                     APPROACH FOR PHASE II(A)  STUDY
RATIONALE

     Nonurban, nonpoint source pollutants are  among the most difficult to
quantify of the many factors which influence the  quality of the nation's
water supply. They are generally introduced into  streams,  lakes and estu-
aries as a result of runoff from rural areas including cropland, grassland,
range, forests and woodlands. The quantity and quality of  rural runoff de-
pend markedly on land use patterns and practices,  and on the climatology
and physiography which characterize a specific watershed.  Such runoff can
contain significant amounts of sediments, pesticides, organic materials,
nutrients and dissolved solids. These pollutants  can represent a significant
portion of the total pollution load of a stream,  and must  therefore be care-
fully considered in the assessment of quality  of  water resources in a given
region.

    The systematic assessment of nonpoint source  pollutants requires not
only data regarding the type of pollutant but  also data on the many fac-
tors which control the rate and manner by which the pollutants are intro-
duced into the water system. These factors include local and basinwide
topography, rainfall rates and frequencies, runoff characteristics, soil
types and prevailing soil conservation practices,  land use patterns, types
of crops and their rotation, fertilizer and pesticide uses, livestock prac-
tices, and other features of agriculture, climate, and physiography.

     As a result of the extensive Data Bank created by MRI during Phase I
of the current study, most of the above factors can be comprehensively
analyzed, for the first time, over the entire  Missouri Basin Region. During
the current Phase II(A) study, we selectively  studied various parameters
of land use, water quality, and other climatic and physiographic factors
to determine the adequacy of .the data base for establishing loading func-
tional relationships due to various pollutant  sources, and to establish on
a preliminary basis the significant factors which can yield meaningful cor-
relations and prediction models. In these analyses, which  required extensive
computer programming, the project staff provided  value judgments and  qualita-
tive inputs which are possible only after a thorough persual of all relevant

-------
data available in published and unpublished form, and from personal contacts
with experts in the field. Thus, we relied extensively on statistical proce-
dures, but these analyses were tempered with insights gained during the pro-
ject period.

     The  scope of the present Phase II(A) study was limited to establishing
the value of the Data Bank and to developing and testing of certain rela-
tionships on a gross scale, in order to verify the validity of the data
base and to develop an understanding of its strengths and limitations.  It
is expected that preliminary functional relationships will lead to the estab-
lishment of cause and effect relationships, which can be translated to plan-
ning and management tools on a local and regional level in the Missouri
Basin, and with suitable modification adopted for national use. This back-
ground is an essential base for continuing study in a suggested Phase II(B)
study aimed directly at development of functions which relate pollutant emis-
sions to land use and land-use patterns, to natural causes and to other
casual factors.

METHOD OF STUDY

     The preliminary data analysis in Phase II(A) required reorganization
of the data base. Separate computer programs were developed to retrieve
selected variables in the data base, and to perform statistical evalua-
tions. The specific methodology used in creating working files of data
and in analyzing these data are discussed in detail in Appendix C.

     The computer program listings are presented in Appendix D.

     An analysis of quality of available data was performed by using stand-
ard statistical techniques. We have routinely used the Statistical Package
for Social Sciences (SPSS), which contains subroutines for analysis of
variance, correlation analysis, bivariate regression analysis, multiple
stepwise linear regression, polynomial regression, factor analysis, non-
parametric tests, and other procedures.

     These computer data processing techniques were used primarily to screen
pertinent variables to develop trends, correlations, and regression equa-
tions. The evolution of, these equations/models was strongly influenced
by the engineering judgment of the.project staff in the selection of key
parameters in thjs modelss-
                                    8

-------
TASKS

     The research program for the Phase II(A)  study included the  follow-
ing tasks in the preliminary analysis of data:

     Task 1 - Evaluation of data for quality and quantity

     Task 2 - Identification of the interrelationships among different
                types of variables

     Task 3 - Preliminary selection and modification of relevant  models

     Task 4 - Development of computer programs

     Task 5 - Appraisal of the models for accuracy and utility.

     The protocol of analyses to evaluate source-stream and other rela-
tionships is shown in Figure 1.

Task 1 - Data Evaluation for Quality and Quantity

     An in-depth analysis was performed on both the quantity and  quality
of data available in our existing Data Bank. This analysis required both
statistical techniques as well as engineering judgments regarding the
operations and processes of land use activities, pollutant generation
and transport, and water quality characteristics in the basin. This an-
alysis was conducted to establish confidence levels for use of these
data in subsequent modeling efforts.

     The analysis included evaluation of the quality of land use  and soil
data by comparison of different data sources,  and use of statistical tech-
niques. The quality of hydrological and water quality data was also simi-
larly analyzed.
                                                     4
Task 2 - Identification of Interactions Among Different Types of  Variables

     During this task, stream water quality parameters were analyzed to
generally characterize water quality throughout the basin and develop sig-
nificant relationships between water quality parameters. Similarly, basin
characteristics (land use, topography, cropping patterns, etc.) were analy-
zed to develop correlations between individual basin characteristics, and
additionally, to develop a basinwide characterization of the Missouri River
Basin.

-------
                       Reduced Station
                       Master File
                       (1969 Data)
                         Screening of
                         Parameters
                         Bivariate
                         Correlation
                         Analysis
                          Multiple
                          Regression
                          Analysis
Figure 1.   Protocol of  statistical evaluation of data
                               10

-------
     The land use and hydrologic data were adequate while data available
on soil and water quality were extensive but contained significant voids.
However, for developing basinwide characterizations of selected variables,
the reliability of available data was shown to be good.

Stream Water Quality Characterization - Spatial water quality changes along
the Missouri River and its tributaries were characterized at selected moni-
toring stations in terms of concentrations of pesticides, nutrients, BOD,
dissolved oxygen (DO), suspended sediment, chloride, alkalinity, and pH.

Basinwide Characterization -  The basin characteristics were evaluated
in terms of land use, soil type, cropping pattern, topography, rainfall,
irrigation, livestock, and chemical application (fertilizers and pesticides).

     Several studies in the literature have related land use practices
                         1 9 /
with pollution potential.iif/ However, such definitive studies are lacking
for the Missouri River Region or for a basin approaching this size. Use
of the MRI Data Bank to produce relationships between pollution potential
and land use patterns in the basin was a significant part of Phase II(A).

Correlations - Using standard statistical correlation and multiple regression
techniques, the following relationships were evaluated:

     • Water quality versus watershed characteristics.

     • Pollutant loading rates versus watershed characteristics.

Task 3 - Preliminary Selection and Modification of Relevant Models

     The preliminary development of models was based on the use of cor-
relation techniques to identify the interrelationships among parameters
of water quality and hydrology, and those of climatological, physical,
and economic characteristics of each watershed. The next step was regres-
sion analysis to identify the significant parameters affecting the quan-
tity and nature of pollutants.
_!/  Bradford, R. R., "Nitrogen and Phosphorus Losses from Agronomy Plots
      in North Alabama," Env. Prot. Tech. Series, EPA, 660/2-74-005 (April
      1974).
2/  Waldon, A. C., "Pesticide Movement from Cropland into Lake Erie,"
      Env. Prot. Tech. Series, EPA, 660/2-74-032 (April 1974).
                                   11

-------
     The preliminary models of Phase II(A) address the following needs:

     1.  To  quantify the  influence of existing nonurban, nonpoint sources
on the pollution  loads of the Missouri River and its tributaries, includ-
ing the effects of natural variations in the source parameters.

     2.  To  evaluate the  technical aspects of changes in nonurban nonpoint
source characteristics as a result of local, regional, or basinwide prac-
tices.

     3.  To  evaluate the  accuracy and adequacy of the available relevant
data by using these data  in the models and assessing the quality of the
results.

     The input to the systems models consists of the Phase I Data Bank
and system constraints imposed by the program user. The constraints repre-
sent user interaction with the program, which enables a variety of techni-
cal alternatives to be evaluated.

     Using presently available models, we initially chose those factors
which may be expected to  be related to runoff and its quality, broke them
down into their simplest  components, and chose factors having the least
interdependence. This task required a knowledge of hydrologic and hydraulic
principles,  and knowledge of behavior of various pollutants, both in soil
and in water. Statistical methods were applied to determine those factors
that are most significant in causing pollution from nonpoint sources. The
physical and economic factors are, in statistical terms, the independent
variables that are to be  associated with the concentration of pollutants
in the river, which are the dependent variables.

     A set of independent variables which are actually independent of each
other would  be preferable, but in our investigation this does not seem
possible. There are very  few variables which are truly independent, e.g.,
soil, groundcover and channel slopes may be affected by the amount of rain-
fall generally available. Thus, topographic and climatologic variables
are not independent of each other; furthermore, fertilizer and pesticide
uses are mutually related to agricultural practices.

     Considering the interdependence of these variables, we used the fol-
lowing analytical procedures:

Correlation  - We investigated the correlation of hydrologic and water quality
parameters with physical  and land use factors. Correlations between various
water quality parameters  and among  hysical and land use parameters were
also studied.
                                    12

-------
     We correlated, for example, BOD^ and livestock;  BOD5 and crop acre-
age; turbidity and soil type; fertilizer use and nutrient (N and P) con-
centration; DO and livestock; crop acreage and pesticide use, stream flow
and precipitation, etc. Correlation coefficients and  variances were ob-
tained.

     For statistical analysis of data we utilized the Statistical Package
for the Social Sciences (SPSS), in addition to the Scientific Subroutine
Package (SSP) developed for use on IBM System/360. The former is useful
in developing display tables of relationships between two or more vari-
ables, bivariate correlation analysis, partial correlation,  and multiple
regression analysis. For example, the subroutine CONDESCRIPTIVE will eval-
uate nine statistics, viz., mean, standard deviation, standard error, vari-
ance, kurtosis, skewness, range, minimum and maximum. PEARSON CORR subrou-
tine computes linear, paired correlations, and can compute means and stan-
dard deviations of the variables listed. The subroutine REGRESSION will
generate linear multiple correlations of two or more  variables, either
stepwise regression, or multiple linear regression. The statistics avail-
able for this subroutine are mean, standard deviation, and correlation
coefficients.

Bivariate Regression - Initially, we developed two parameter regressions,
i.e., bivariate regression analysis for pairs of parameters, to evaluate
the effect of several parameters on water quality.

     These analyses involved water quality parameters including DO (dis-
solved oxygen), turbidity, BOD (biochemical oxygen demand),  P (phosphorus),
NOg (nitrate), and DS (dissolved solids); and land use parameters includ-
ing cattle, hogs, cattle and hogs, livestock (cattle, hogs,  sheep, and
chickens, converted into equivalent cattle), nitrogen and phosphate ferti-
lizer applied, cropland and irrigated land.

     The regressions were based on actual data, and on data transformed
into natural logarithms. Data were fitted to a straight line form, Y =
A + BX, where Y is a water quality parameter, X is a  land use or physio-
graphic parameter, and A and B are regression coefficients.

Multiple Regression Analysis - Multiple regression techniques were used
to relate water quality parameters to physical, climatic and economic
characteristics. The equation employed was based on the findings of cor-
relations among variables and available correlations  such as those de-
veloped by the U.S, Geological Survey. A stepwise multiple regression
program was used to calculate the regression equation, the standard
error of estimate, and the significance of each basin parameter. The
                                  13

-------
calculations were  repeated,  omitting the  least significant basin parameter
in  each  calculation until only the most significant parameters remained.

     Four water  quality parameters were used for this study.  These were:

     a.  Turbidity x flow

     b.  BOD

     c.  Nitrate

     d.  Turbidity

     Calculations  were repeated on data transformed into natural logarithms.
Land use parameters selected for multiple regression analyses included crop-
land,  irrigated  land, cattle plus hogs, nitrogen and phosphate applied,
rainfall, slope, and cover factor. A linear regression equation was used
to  fit the data. This equation is of the general form,

                                 n
                      Y± = Ci + ^ AM * X±f.


Task 4 - Development of Computer Programs

     A listing of  the computer programs developed during the study is shown
in Appendix D. These include both the new data files created in a form that
increases the utility of the data base at least cost, and the programs for
statistical data analysis.

Task 5 - Appraisal  of the Models for Accuracy and Utility

     The data analysis resulted in several models that relate various water
quality parameters  to land use  parameters. Their significance and reliabil-
ity within the constraints of the quality of the available data was appraised.
                                   14

-------
                               SECTION V

                                RESULTS

     Data analysis was aimed primarily at establishing that  the  data  in
our Data Bank are adequate to develop correlations and regressions  among
variables of selected water quality,  land use,  and physiographic param-
eters in the basin. In this section,  we present the results  of data analy-
sis in the same sequence which enabled us to develop multiple regressions
on selected variables. An evaluation  and discussion of the results  will
be presented in Section VI.

     Data from the Station Master File were used as the primary  source
in the data analysis. In this data file, land use, physiographic, and water.
quality data are listed for 233 watersheds, each being represented  by a
water quality monitoring station. These watersheds were assigned to 13
subbasins according to the classification of the U.S. Geologica'l Survey
Office of Water Data Coordination (OWDC). The entire Missouri Basin was
covered by these subbasins:  the northwestern portion of Missouri,  the
northern portion of Kansas, all of Nebraska, the western half of Iowa,
the southwestern tip of Minnesota, most of North and South Dakota,  north-
eastern Colorado, and most of Wyoming and Montana.

     The latest land use data in the  data file were reported for 1969,
while water quality data were compiled for 3 years—1968, 1969,  and 1970.
After physical examination of the nature and extent of these data,  it was
decided to analyze 1 year's data (1969), which are available in  terms of
most parameters being analyzed. Consequently, the data analysis  presented
in this report is limited primarily to 1969 data unless otherwise stated.

BASINWIDE LAND USE CHARACTERIZATION

     Basin land use data summarized for each subbasin and the entire basin
are presented in Table 1. Mean values of various land use characteristics
of the basin are presented in Table 2.
                                   15

-------
                    Table 1.   SUMMARY OF LAND USE DATA BY SUBBASINS FOR 1969


Subbasin
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
Total
land
(miles2)
48,786
32,927
28,568
32,184
75,147
32,493
22,225
19,095
74,427
14,372
16,174
29,720
35,245
Farm-
land
(miles2)
41,361
31,385
27,201
21,397
71,187
30,958
21,418
16,653
64,420
10,819
8,238
26,275
18,144
Total
cropland
(miles2)
27,286
21,316
12,809
3,507
34,672
24,681
5,232
1,363
23,974
3,680
1,739
2,421
2,420
Harvested
cropland
(miles2)
13,684
11,915
6,323
1,947
21,797
16,498
3,347
788
11,899
1,738
984
1,235
1,308
Grazing
land
(miles2)
8,739
2,290
973
460
4,797
3,142
705
246
1,683
155
358
305
491
Other
cropland
(miles2)
4,865
7,027
5,525
1,095
^ 8,076
5,042
1,184
329
10,394
1,786
399
878
581
Other
farmland
(miles2)
9,249
11,612
16,015
19,017
37,435
6,111
16,226
14,892
40,232
7,206
6,976
24,033
16,342

No. of
cattle
4,465,830
2,287,146
1,598,127
1,019,345
5,658,978
3,021,364
1,057,727
406,489
1,715,289
231,951
401,917
615,602
701,450
Total    461,363    389,456    165,100
93,463
24,344
47,181    225,346   23,181,215

-------
                                   Table 1.  (Concluded)


Subbasin
No.
31
32
33
34
35
36
37
38
39
40
41
42
43

No. of
hops
3,700,371
989,528
246,737
63,931
3,392,060
2,658,911
203,361
13,576
144,369
20,742
22,407
17,404
47,280

No. of
chickens
3,540,841
1,524,760
376,156
484,931
4,198,840
4,285,791
224,704
54,415
477,043
44,829
207,175
62,097
211,605

N
(tons)
548,215
421,126
130,878
42,774
515,411
201,003
20,153
3,392
47,414
4,529
10,586
9,185
32,972

P04
(tons)
309,882
103,878
46,363
15,241
190,424
132,937
10,434
1,690
63,542
5,914
13,816
10,491
24,548
Fertilizer
(N + P04)
(tons)
858,097
525,004
176,821
58,015
705,835
333,940
30,587
5,082
110,956
10,44.3
24,402
19,676
57,520

Herbicide
(tons)
8,808
4,223
1,796
392
7,619
7,008
382
105
2,184
358
124
322
454

Insecticide
(tons)
3,487
1,955
902
232
3,568
1,788
128
44
512
47
16
120
182
Total    11,518,677   15,693,187  1,987,218    929,160    2,916,378
33,775
12,981

-------
                       Table 2.   ANNUAL LAND USE LOADING IN MISSOURI BASIN, STATION MEAN VALUES
oo


Subbasin
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
Mean
Median

Cattle
(No./
miles2)
98.142
73.525
51.089
28.109
117.163
94.329
51.009
22.390
30.883
16.157
24.746
18.9-28
18.608
62.869
54.1

Hogs
(No./
miles2)
79.309
33.206
8.290
1.999
77.075
84.832
12.679
0.737
1.829
1.496
1.109
0.339
1.282
34.211'
8.7

All livestock
(as cattle)
(No. /miles2)
120.005
83.304
54.420
32.221
139.034
120.644
55.213
26.004
33.445
17.050
26.914
22.978
21.546
73.929
57.8

Nitrogen
(tons/
miles2)
11.875
13.892
4.241
1.253
15.175
5.568
1.089
0.195
0.504
0.323
0.641
0.231
0.956
7.084
3.5

Phosphate
(tons/
miles2)
6.466
3.755
1.451
0.471
4.444
3.746
0.565
0.097
0.652
0.422
0.837
0.187
0.633
2.698
1.3

Cropland
Irrigated
land
(miles2/ (miles2/
miles2)
0.580
0.665
0.441
0.098
0.572
0.770
0.270
0.083
0.307
0.282
0.109
0.047
0.063
0.410
0.44
miles2)
0.003
0.062
0.056
0.043
0.056
0.001
0.014
0.008
0.004
0.010
0.053
0.015
0.029
0.039
0.02

Annua 1
rainfall
(in.)
38.802
29.132
19.026
15.247
24.830
21.848
18.607
14.860
16.293
12.403
13.668
12.053
12.048
22. 246
21.4

-------
     These values are computed for each watershed by dividing the value
of the land use parameter by the area of watershed represented ry the water
quality monitoring station. Thus, Station No.  1000016 in subbasin 31 has
a total land area of 3,829 miles2, and includes 391,398 cattle.  The com-
puted cattle density for this station is 102.219 cattle/mile2.  An average
of all such density values for subbasin No. 31 is 98.142 cattle/mile2,  as
shown in Table 2. On the other hand, the average cattle density computed
from Table 1 is 91.539 cattle/mile  which is significantly different from
the station mean value of 98.142.
                                                                       t

     Figures 2 through 10 show the distribution of various land use  char-
acteristics in the Missouri Basin. Each subbasin in these figures is la-
beled by a number (31 through 43) along with the value of the measured or
calculated parameter.

     Figures 2 and 3 show the distribution of cattle and hogs,  respec-
tively, in the Missouri Basin. Subbasins 31, 32, 35, and 36 contain  the
largest numbers of cattle and hogs. Among these four, subbasin 31 has the
largest number of cattle and hogs of all the subbasins.

     Figures 4 and 6 show the densities of cattle and hogs/mile2 in  each
subbasin. In the case of cattle, subbasins 31, 32, 35, and 36 again  record
highest densities, with subbasin 35 showing the highest cattle density
(117 cattle/mile ) of all subbasins. In the case of hogs, subbasins  31,
35, and 36 show higher densities than the basin average, with subbasin 36
reporting the highest density (85 hogs/mile ).

     Figures 7, 8, and 9 show the distribution of fertilizer (nitrogen
and phosphate), herbicide, and insecticide uses, respectively,  in the basin.
In each case, the four subbasins 31, 32, 35, and 36 show values above basin
average. Subbasin 31 has the largest use of fertilizer and herbicide, while
subbasin 35 has the largest insectide use.

     Figure 10 shows the loading of fertilizer per unit area (tons/mile )
Subbasins 31, 32, and 35 show values greater than the basin average, sub-
basin 35 showing the highest loading value of all subbasins.

BASINWIDE WATER QUALITY CHARACTERIZATION

     Table 3 presents water quality characteristics in terms of selected
parameters in each subbasin. These parameters include coliforms, dissolved
oxygen (DO), biochemical oxygen demand (BOD), turbidity, total filterable
residue (TFR) or dissolved solids, nitrate (NO-j), and total phosphorus
(P). Averages for the entire basin were also computed and presented  in
Table 3.
                                   19

-------
                                                  CANADA
Ni
O
                       (0.70)
                        WVOM/NG
S. DAKOTA"\(3.02)
                                                  NEBRASKA     35
                                                             (5.66)
                                Figure 2.  Distribution of cattle in Missouri Basin
                                            (millions of cattle)

-------
                    CANADA
Figure 3.  Distribution of hogs  in Missouri Basin
             (millions of hogs)

-------
                                                    CANADA
IV)
                                                   S. DAKOTA S (94)
              Figure 4.  Distribution of cattle per  square mile in Missouri Basin,  Station Mean Values
                          (basin average = 63)

-------
                                                                      92°
Figure 5.  Distribution of cattle per square mile in Missouri Basin

-------
                                    CANADA
Figure 6.   Distribution of hogs per square mile in Missouri Basin, Station Mean Values
             (basin average =34)

-------
                       CANADA
                                                   31
                                                  (858)

                                                      MISSOURI
Figure 7.   Fertilizer use (N and  PO^) in Missouri Basin
             (values shown are 1,000's of tons)

-------
                                                  CANADA
N)
                                     Herbicide use  (tons) in Missouri  Basin
                                       (values shown in 1,000's of tons)

-------
                                                 CANADA
FO
                                                S. DAKOTAS (1.8)
                           Figure 9.   Insecticide (tons) use in Missouri Basin
                                        (values shown are in 1,000's of tons)

-------
                                                CANADA
00
                                                             36
                                                S. DAKOTA S  (9)
                                                NEBRASKA     35
                                                                     f\
       Figure  10.  Fertilizer (N and P) use in Missouri Basin (tons/miles ),  Station Mean Values

                    (basin average = 10 tons miles^)

-------
                                        Table  3.  WATER QUALITY IN MISSOURI BASIN
ro
\o


Subbasin
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
_
Coliforms
(No. 7100 ml)
32,799
5,509
386
18,847
201,326
3,607
5,542
10,029
47
6,675
.
5,092
1,332

DO
(me/ft)
8.181
9.253
9.090
6.492
8.998
9.594
9.000
8.048
9.108
8.699
-
8.919
10.120

BOD
(ms/n
4.075
4.451
2.991
4.184
15.646
7.462
4.941
5.574
1.158
1.802
-
3.260
1.598

Turbidity
JU
216.936
272.283
115.361
72.481
588.911
80.363
629.695
449.586
39.778
97.400
-
1,262.149
20.272
Total filter-
able residue
(DS) (mg/A)
312.747
623.855
724.145
483.395
672.784
643.272
1,076.400
1,657.689
879.055
567.904
362.306
1,687.281
407.031

N03
(mg/jfc )
4.047
3.710
3.433
1,776
5.421
1.643
4.330
3.234
0.578
0.313
0.539
1.410
0.717

P, total
Cms/A)
0.281
0.092
-
-
2.369
0.944
4.833
-
0.045
-
-
0.167
0.228
Number of
water quality
stations
28
41
20
27
32
18
9
9
5
4
5
11
24
       Mean for

         the basin     24,266      8.669
6.819
327.090
776.758   2.672
1.169
(233)

-------
     Specific pollutant loading rates were computed for five selected
water quality parameters, viz., DO, BOD, N03, P, and DS. These rates,
expressed  in units of pounds of pollutant per square mile of land per
day, are presented in Table 4 for each of the subbasins and the whole
basin. These values were computed for each  water quality monitoring
station from the pollutant concentration (mg/liter) and stream flow
(ft-5/sec)  and land area of watershed (mile^) represented by that sta-
tion. The  number of stations .selected for each subbasin are also pre-
sented. "Loading rates" for DO are included in Table 4. DO and BOD
are usually inversely related. However, DO is also related to other
stream parameters, e.g., phytoplankton, sediments, and temperature.

     Table 5 presents a summary of both water quality concentrations and
pollutant  loading rates for seven parameters. These parameters include
DO, BOD, turbidity, dissolved solids, nitrate, total phosphorus, and coli-
forms. The units for each of the parameters are also shown in Table 5.

     Figures 11 through 17 show the distribution of selected water quality
parameters in the basin. These parameters include concentrations of dis-
solved oxygen (Figure 11), nitrate (Figure 12), BOD (Figure 13), total
coliforms  (Figure 14), turbidity (Figure 15), and dissolved solids (Fig-
ure 16). Figure 17 shows BOD in pounds per day per square mile in the
basin.

     Subbasin 35 shows the highest concentration for all pollutants except
turbidity. Subbasins 31, 35, and 38 show lower dissolved oxygen concentra-
tions than the basin average. Nitrate concentration is higher than the
basin average in subbasins 31, 32, 33, 35, 37, and 38. BOD concentration
is higher  in subbasins 35 and 36, while total coliforms concentrations
show higher values in subbasins 31 and 35 than the basin average. Turbidity
values show that subbasins 35, 37, 38, and 42 have above average values.

STREAM WATER QUALITY CHARACTERISTICS

     Water quality data for 1969 in the Missouri River and other streams
in the basin were analyzed in terms of DO, BOD, N03, and total P, as a
function of location of water quality monitoring stations. Eleven graphs
(Figures 18 through 28) are presented for these parameters as well as
flowrate in the streams. Some of the streams which were also analyzed
but not graphically presented showed generally good water quality in terms
of the above parameters, but did not contain sufficient data for analysis.
The following streams were analyzed in detail:

     Missouri River
     Yellowstone River
     Bighorn River


                                   30

-------
    Table 4.  POLLUTANT BURDEN IN MISSOURI BASIN (Ib/mile2/day)
Subbasin
No.
31
32
33
34
35
36
37
38
39
40
41
42
43
Average for
the basin
DOS./
168.103
71.855
7.933
25.149
68.055
15.856
0.970
11.554
-
16.249
-
18.22
180.619
88.548
BOD
149.900
46.418
0.482
7.969
75.771
9.989
0.311
7.767
-
3.366
-
6.66
21.858
40.091
N03
113.065
29.186
1.749
7.125
22.194
28.611
0.201
3.837
0.071
0.424
1.214
1.658
8.154
23.221
P DS
54.624 6,665
0.735 2,737
260
173
6.351 2,070
1,380
108
1,042
116
589
1,107
0.633 1,602
4.125 4,028
12.136 2,493
a/  DO is included here as a water quality parameter rather than as a
      pollutant.
                                  31

-------
       Table 5.  SUMMARY OF WATER QUALITY AND POLLUTANT BURDEN
                       IN MISSOURI RIVER BASIN
   Parameter

DO

BOD

Turbidity

Dissolved solids

Nitrate (N03>

Total phosphorus

Coliforms
    Water quality
    concentrat ions
    Mean
concentrat ion
                                               Pollutant burden
                Unit      Mean value       Unit

     8.70      mg/liter       88.5    Ib/miles2/day

     6.82      mg/liter       40.1    Ib/miles2/day

   327           JU             0.58   JU/miles2

   777         mg/liter     2,464     Ib/miles2/day

     2.67      mg/liter       23.2    Ib/miles2/day

     1.17      mg/liter       12.1    Ib/miles2/day

24,266        No./100 ml
                                   32

-------
                                                  CANADA
U3
CO
                                                 S. DAKOTA ^9.59)
                   Figure  11.  Dissolved oxygen (ing/liter) in streams of Missouri Basin
                                (basin average = 8.67)

-------
                          CANADA
                          NEBRASKA    . 35
Figure 12.  Nitrate (mg/liter) in streams of Missouri Basin
             (basin average = 2.67 mg/liter)

-------
                                                CANADA
                                                NEBRASKA   ,35.
Oi
                         Figure 13.  BOD (mg/liter)  in streams of Missouri Basin
                                       (basin average = 6.82 mg/liter)

-------
                            CANADA
                           S. DAKOTA M3,600)
Figure 14.  Total coliform (No./lOO ml)  in streams of Missouri Basin
             (basin average = 24,266)

-------
                    CANADA
                    S. DAKOTA S  (80)
Figure  15.  Turbidity  (JU) in streams of Missouri Basin
             (basin average = 327)

-------
                                                   CANADA
U)
00
                                                   NEBRASKA    , 35
                         Figure 16.  Distribution of dissolved solids in Missouri Basin

                                      (basin average = 777 mg/liter)

-------
                                                   CANADA
co
                                                   S.DAKOTA'S (10.0)
     Figure 17.   Distribution of BOD (Ib/day) per square mile in Missouri Basin,  Station Mean Values
                   (basin average = 40 lb/day/mile2)

-------
\JL
10
8
€
X
t«2 O
o
•t
J
_o
u_

4
2

12
11
10

1
- . 9
0
Q



8
— 7

6
5
1 4
^
CO „
- O 3
Z
Q
O .
CO
2
1
0
1.0
0.8<
^
- ^0.6
S
H-


0.4
0.2
(
<
500
1000
                                                                         1500
                                       River Distance from its Origin, Miles




Figure  18.  Water  quality along the main  stem of Missouri River
2000

-------
14,000 r
12,000
10,000
 8,000
 6,000
4,000
2,000
0L
12
11
         0
         Q
10
             >  2

             Q
             o
             CQ

              CO
             o
                                                     A,
                                          J_
                                         J_
                                                               _L
                                100       200        300       400

                                  River Distance from its Origin, Miles
                                                                     \
                                                             500
    Figure 19.  Water quality along Yellowstone  River

-------
to
                             7000
                              6000
                              5000
                              4000
                              3000
                              2000
                              1000
   15
                                                                                                                   90.8
   14
-  13
                                     12
 O
 Q

-  11
   10
                                 OL   8>-    0
  ?


  Q

  O
  CO


  co

ko
                                50                    100


                                        River Distance from its Origin, Miles
                                                                                                             150
                                                                                            200
                                     Figure  20.  Water quality  along Big Horn  River

-------
               300
CO
               200
               100
                                                 40          60          80         100         120         140
                                                     River Distance from its Origin, Miles
                              Figure 21.  Water quality along Belle Fourche River

-------
60r        12r
50
40
30
20
 0L
                          '              ^-—""~
                                              River Distance from its Origin, Miles
                       Figure 22.   Water quality along White River

-------
Oi
         100
          80
          60
          40
          20
                     121-
10
                       1   DO
o
I  6
IE
                    u
                        \
     \Flow

       \
         \
                           BOD	
                                                                  _L
                                                                             _L
                                             50
                                             100                    150

                                    River Distance from its Origin, Miles
                                                                                                             200
                                                                                                              250
                                      Figure  23.  Water quality along Niobrary  River

-------
7000 r   7r  17r
6000 -   6 -
5000
-   5
4000
-   4

      6°
3000
 2000
-   2
 1000
    >L   0
                                     50                    100                   150
                                            River Distance from its Origin, Miles
                                                                                                 200
      Figure 24.   Water quality along Big  Sioux River

-------
 3000 r   12 r   3r
 2000
I
 1000
   O1-
     -   11 -
-10-
                                       100
                                                         200
                                           River Distance from its Origin, Miles
300
400
                       Figure 25.   Water quality along North Platte River

-------
                1200
00
                1000
                 800
                 600
o
O
                 400
                 200
             17 i-
             15  -
             13 -
 CO
i
d
Q
                                                               100                      200
                                                            River Distance from its Origin, Miles
                                                                                       300
                             Figure  26.   Water quality along South  Platte River

-------
                                                                                                      Missouri River
       6000 r
                  11
       4000
   10
•p-
vO
o
Q
       2000

                                                   100
                                                            200                     300

                                                    River Distance from its Origin, Miles
400
                                        Figure  27.   Water quality along Platte River

-------
Ui
o
                                      13
                               1500
-  12
                                   h  11
                               1000
h  10
                                    O
                                500
                                   l-   6
     h   5
       o
       O
       CO
        •t
       CO
      -03
                                        [-   1
     L-   0
                                                                                                                       u

                                                                   100                    200                   300

                                                                            River Distance from its Origin, Miles
                                                                                                  400
                                     Figure 28.   Water quality along Republican River

-------
     Belle Fourche River
     White River
     Niobrara River
     Big Sioux River
     North Platte River
     South Platte River
     Platte River
     Republican River

     The following streams were not analyzed in detail due to  lack  of  adequate
data, but were found to possess generally adequate water quality:

     Milk River
     Powder River
     Cheyenne River
     James River
     Elkhorn River
     Loupe River
     Lodgepole Creek
     Saline River
     Smoky Hill River

     The following streams were also not analyzed in-depth due to  lack
of adequate data, but available data appear to indicate that these  streams
are generally of poor water quality, and that further acquisition of data
is warranted:

     Elkhorn River (high N03)
     Kansas River (high N©3 and DS)
     Grand River (high NC>3, very high coliforms)
     Moreau River (high coliforms, low DO)
     Osage River (high N03, coliforms, low DO)

     A brief discussion of water quality in the Missouri River and  other
streams follows.

Missouri River

     Five water quality monitoring stations were selected on the main  stem
of the Missouri River. These were located at Canyon Ferry, Montana  (Station
No. 1000001); Bismarck, North Dakota (1000006); Omaha, Nebraska (1000011);
Kansas Gity, Kansas (1000016); and Hermann, Missouri (1000021). Water  quality
data for these widely separated stations are presented in Figure 18.
                                   51

-------
     From Figure 18, it is evident that BOD, total phosphorus, and nitrate
concentrations increase downstream, while DO concentration decreases.  This
trend appears to be more pronounced beyond Omaha, Nebraska (Station No.
1000011). From the flow data, it is evident that both the Platte and Kansas
rivers contribute to the flow in the Missouri River on the downstream  end
of the nearest water quality station.

     The increase of pollutant  concentrations at and below these two sta-
tions indicate that both point  sources and nonpoint sources are major con-
tributors to  stream pollution.  The point  source contributions include
municipal discharges from urban wastes in the area--0maha, St. Joseph,
and Kansas City—as well as other numerous smaller townships within the
drainage basin upstream of the  two stations. The nonpoint source contri-
butions appear to be primarily  due to feedlots, croplands, and other ag-
ricultural activities which are concentrated in the southeastern portion
of the basin. However, the amount of pollution contributed by individual
point and nonpoint sources cannot be computed from the current Data Bank
due to a lack of data for point sources.

Yellowstone River

     A major tributary of the Missouri River, the Yellowstone River origi-
nates near Yellowstone Lake in  northwest  Wyoming, flows northeasterly through
Montana, and joins the Missouri River just beyond the Montana-North Dakota
line. The City of Billings is a major metropolitan area located on the
Yellowstone River, which has three major  tributaries, viz., Bighorn River,
Tongue River, and Powder River. Shoshone  River is a tributary stream to
Bighorn River. Two mountain ranges cross  the Yellowstone River Basin.  These
are the Absaroka Range and Bighorn Mountains. Yellowstone National Park
is a major recreational area.

     The Yellowstone River main-stem drainage basin has a total land area
of 27,263 miles2, of which only 11% is cropland, and only 5% is harvested
cropland. It has 692,250 cattle, 35,800 hogs, and 188,250 chickens, the
density of livestock being 34/mile2. Fertilizer application rate averages
only about 1.3 tons/mile^/year.

     Figure 19 shows the changes in DO, BOD, and NOo concentrations in
the river at various stations located along the path of the river. In the
graph, the variation of flowrate at these stations is also shown. Although
some stations do not contain sufficient data for water quality parameters
shown, it is clear that the concentration levels for N03 and BOD are very
low, and DO levels quite high,  indicating a relatively high quality stream.
The turbidity levels are also very low (often less than 100 J-U). Thus, the
relatively low levels of land uses appear to produce little degradation in
stream water quality.


                                    52

-------
Bighorn River

     Bighorn River is a tributary to Yellowstone River.  The water  quality
in Bighorn River is also generally good as  indicated by  DO and  NC>3  (Figure
20). The BOD concentration for one station  indicates an  unusually  high
value (90.8 rag/liter), which is not explained by other water  quality  data.
Land uses in the drainage basin include an  extremely low level  of  cropland
(5.7%), about 20 livestock/mile2, and about 1.5  tons of  fertilizer used
per square mile. These values are similar to those of the Yellowstone
River Basin.

Belle Fourche River

     A tributary to the Cheyenne River, which joins the  Missouri River
through Lake Oahe in the central plains of  South Dakota, Belle  Fourche
River flows through Wyoming and South Dakota. Less than  10% of  the land
draining into Belle Fourche River is cropland, and relatively low  density
livestock operations prevail in the basin.  Livestock data for 1969  show
that there are 130,940 cattle, 4,360 hogs,  and 17,910 chickens  among  other
                                                      o
livestock, with an average density in excess of  26/mile  . There is  a  pro-
gressive increase in the use of commercial  fertilizer in the  basin in the
downstream direction, with an average application rate of nitrogen and
phosphate fertilizer of about 0.3 tons/mile /year.

     Figure 21 shows the distribution profile of NO-j, BOD, and  DO  con-
centrations in milligrams per liter, and flowrates, in GFS, at  various
stations on the river. Although data are limited for BOD and  DO, the
rapid increase in nitrate concentration, particularly below the town  of
Belle Fourche, indicates deterioration of water  quality  in the  stream.
This increase of pollutant concentration, in spite of increased flowrate,
suggests that the possible sources of pollution  are feedlots  near  the
stream.

White River

     The White River is a tributary of the  Missouri River, originating
in the Pine Ridge Range of northwest Nebraska and flowing through  the
south central plains of South Dakota. The drainage basin is essentially
rural and is sparsely populated. Its dedication  to agriculture  is  reflected
by the farmland (nearly all of the total land) and cropland (about 20%
of the total land), and in its livestock operations, with 341,000  cattle,
53,300 hogs, and 61,000 chickens on an area of only 9,340 miles2.  This
represents a livestock density of over 50/mile2. The use of commercial
fertilizer is about 0.5 tons/mile2/year.
                                    53

-------
     A profile of concentrations of NC^, BOD, and DO, in milligrams per
liter at various water quality monitoring stations is presented graphically
in Figure 22, which also shows the flowrate at these stations. The graph
indicates that both BOD and N03 concentrations increase with a decline in
DO concentration along the river. Thus, there is a gradual decline in water
quality, due most probably to the high density of livestock in the basin.

Niobrara River

     A tributary of the Missouri River, the Niobrara River flows mostly
through the northern part of Nebraska, which is a very sparsely populated
region. There are few major urban centers intercepting or draining into
this river, but a relatively high density of livestock (75/mile2) exists
in the basin. A high proportion of land use is related to cropland (20%)
and the overall commercial fertilizer application rate is moderate (2.1
tons/mile^/year). In spite of intensive agricultural operations in this
river basin, there appears to be relatively little degradation of water
quality, as reflected by DO concentration in Figure 23. Nitrate data
are not available to assess the effect of fertilizer application, but there
appears to be an increase in BOD concentration in the stream from the data
for BOD at Station No. 4100016 (3.25 mg/liter).

Big Sioux River

     This river, a tributary of the Missouri River, flows southerly through
South Dakota and Nebraska, with its confluence near Sioux City. Sioux Falls,
Brookings, and Watertown are major cities on the river. The river basin is
urbanized with many small towns and villages scattered in the basin. Its
land use pattern is predominantly agricultural, with high concentrations of
livestock and crop production. The proportion of cropland is about 80%,
most of which is devoted to harvested cropland. The number of cattle, hogs,
and chickens produced in the basin is high; cattle, 852,000; hogs, 85,000;
and chickens, 1,450,000; with the average density of livestock being about
404/mile2. The use of commercial fertilizer is also quite high, averaging
about 11 tons of nitrogen and phosphate fertilizer per square mile per year.
The use of herbicides and insecticides in the basin is also higher than
the average values for the entire Missouri Basin.

     The water quality changes in the river are presented graphically in
Figure 24. The graphs show rapid increases in nitrate and BOD, and a de-
crease of DO in the river. The water quality degradation in the stream is
consistent with the high use of fertilizers, crop harvesting, and feed-
lot operations in the Basin.
                                   54

-------
North Platte River

     The Platte River Basin comprises a large drainage  area in  three  states
of Wyoming, Colorado, and Nebraska, with several streams  joining  the  Platte
River, which is a major tributary of the Missouri River.  Among  these  streams,
the North Platte is the longest, originating in the Rocky Mountains in nor-
thern Colorado and flowing northerly in Wyoming in the  foothills  of Medi-
cine Bow Mountains, easterly along the foothills of the Laramie Mountains
and in the great plains in Nebraska, and finally joining  the Platte River
near North Platte, Nebraska. The drainage basin is well populated, but the
predominant land use is agriculture. However, the intensity of  agricultural
use is relatively low: cropland (less than 10%); livestock  (427,000 cattle,
21,500 hogs, 24,600 chickens); density of livestock (32.I/mile2). The use
of commercial fertilizer is also moderate (2.2 tons/mile2/year).

     Figure 25 shows water quality in the North Platte  River at various
stations. The NC>3 and BOD concentrations appear to be low,  while  DO concen-
tration is relatively high, indicating that the stream  water quality  is
generally satisfactory.

South Platte River

     The South Platte River also originates in the Rocky  Mountains, passing
through Colorado and Nebraska, before joining the Platte  River  near the
City of North Platte. The major metropolitan area located on the  South
Platte River is Denver, Colorado; several other towns are scattered along
the river.

     The drainage basin is primarily agricultural, with cropland  (327o)
and livestock operations (105/mile ) dominating its land  use. There is
also a relatively high rate of commercial fertilizer use  in the basin (5.5
tons/mile2/year).

     These land use characteristics are reflected in the  water  quality
 'hanges in the stream, as shown in Figure 26. BOD values  are usually
aigh in the middle segment of the river, while nitrate  concentration  is
shown to be increasing. Thus, the general water quality of  the  stream ap-
pears to be on the decrease.

Platte River

     The Platte River, a tributary to the Missouri River, flows through
the central plains of Nebraska and joins the Missouri River at  Plattesmouth,
Nebraska. From North Platte to Plattesmouth, the Platte River  is  about 340
miles in length and becomes increasingly urbanized as it  progresses towards
its confluence with the Missouri River.

                                     55

-------
     The Platte River  Basin has one of the most intensive agricultural
land uses  found anywhere  in the Basin. It has about 65% of total land
under cultivation  for  crops, corn being the largest harvested crop.
There are more than  776,000 cattle, 493,000 hogs, and 636,000 chickens
in the  six watersheds  covering the Platte River Basin, with a livestock
density of more than 300/mile . Commercial fertilizer use is also one of
the most intensive:  142,785 tons of nitrogen fertilizer and 38,041 tons
of phosphate  fertilizer or about 30 tons of fertilizer (nitrogen and phos-
phate)  per square  mile per year. The area is also characterized by a high
application of herbicides (1,422 tons) and insecticides (750 tons) with
an annual  average  rate of insecticide application at about 0.35 tons/mile .

     These intensive land uses are reflected in the water quality changes
at various stations  in the Platte River. Figure 27 shows BOD, NO 3 , and
DO concentration profiles at various stations, along with the flowrate of
the stream. BOD concentration has a pronounced upward slope along the down-
stream  direction,  while DO concentration is shown to be decreasing. How-
ever, N03  concentration does not appear to be increasing in the Platte
River,  although in the Missouri, N03 as well as BOD concentrations are
rapidly increasing.  Thus, from a water quality point of view, there is a
high correlation between  land use, increased concentration of biodegradable
pollutants, and decreased DO concentration in the river.

Republican River

     The Republican  River is a tributary to the Kansas River, which joins
the Missouri  River at  Kansas City. The Republican River originates in east-
ern Colorado  and flows mostly through southern Nebraska before joining the
Kansas  River  near  Junction City, Kansas, through Milford Reservoir.

     The total area  of the watershed draining into the Republican River
upstream of Orleans, Nebraska, is 12,383 miles2, of which more than 50%
is cropland.  There are more than 761,000 cattle, nearly 300,000 hogs, and
320,000 chickens in  the basin. The total livestock density thus averagjea  nc
over Ill/mile2. The  use of commercial fertilizer is also quite intensive,
averaging  about 11 tons of nitrogen and phosphate fertilizer per square
mile of the Basin. The use of herbicide and insecticide chemicals amounts
to about 0.15 tons/mile2/year. Thus, the potential water pollution problems
for intensive agricultural land uses in the basin are significant.

     Figure 28 shows water quality changes in the river basin at various
monitoring stations, in terms of DO, N03, and BOD. In this figure, flow-
rates are  also represented.
                                    56

-------
     The variations of BOD and N03 concentrations show very little upward
trend, while the DO concentration appears to fluctuate significantly,  drop-
ping from a high value of 10.3 near Trenton, Nebraska, to a low value  of
8.3 near McCook, Nebraska. The flowrate increases gradually in the down-
stream direction in the river, increasing rapidly beyond the City of McCook.
This increase of flowrate may have resulted in the dilution of pollutants
and prevented elevated concentrations of pollutants in the river.

     Thus, water quality in the Republican River does not appear to reflect
the intense agricultural operations in the river basin, presumably because
the river has a high capacity to assimilate pollutants.

REGRESSION ANALYSIS

     Statistical techniques were used to determine the degree of associ-
ation between variables and the functional relationships between the vari-
ables. If X and Y are both random variables which follow some unknown  bi-
variate distribution, the degree of association between the two variables
in the distribution pattern is explained by a correlation coefficient  be-
tween X and Y within a given sample space. Regression analysis establishes
the dependency relationship that may exist between X and Y. In a linear
regression analysis, the observed values of Y for given values of X are
fitted to a straight line, using the method of least squares, on an X-Y
plane. Here Y is a dependent variable, while X is an independent variable,
i.e., X is a random variable which determines the changes in Y. Thus,  from
a regression equation, one can predict the size of the change in Y when
a unit change is made in X.

     When more than two variables are involved, the same regression tech-
nique can be applied on an n-dimensional space, when n corresponds to  the
number of variables. When the random variables are selected one at a time
in multiple regression analysis, the method is called stepwise regression.

   jr Both bivariate and multivariate regression analyses were performed
on fHiep'data. The parameters which were assumed to be random and thus "in-
depfSraent:" variables were land use parameters, soil, and climatic factors.
The dependent variables were assumed to be the parameters of water quality,
since the various land use activities of man, and natural geophysical factors,
cause changes in water quality in streams.

Bivariate Regression Analysis

     Basically, bivariate regression analyses have been made between water
quality parameters, including DO (dissolved oxygen), turbidity, BOD (bio-
chemical oxygen demand), P (phosphate), N0-j (nitrate), and DS  (dissolved
                                   57

-------
solids); and land use parameters, including cattle, hogs, cattle and hogs,
livestock (cattle + hogs + sheep + chickens) (livestock populations  were
expressed in units of animal waste generation, with a beef animal as the
unit reference point), nitrogen fertilizer, phosphate fertilizer, and total
fertilizer (nitrogen + phosphate) (in tons per year), cropland (square
miles) and irrigated land (square miles). For water quality parameters,
we have used two types of measurement. One is the pollutant load (e.g.,
pounds per day per square mile, obtained by multiplying stream flow  with
pollutant concentration and dividing by watershed area). The second  type
of measurement is the concentration in milligrams per liter. Land use param-
eters have been converted to a unit area base, e.g., tons of N fertilizer
per square mile of watershed.

     In the subsequent analysis of water pollution due to livestock, it
was found necessary to express livestock populations in terms of an  equi-
valent population, preferably cattle, as cattle wastes represent the
largest livestock wastes in the basin. The weights assigned were found
to be dependent on the specific pollutants being analyzed. Thus, three
sets of weights were chosen for three pollutants, viz., BOD, nitrogen as
N, and phosphorus as ]?205*

     The relative weights assigned to the wastes from cattle, hogs,  sheep,
and chickens were computed from the literature, and are shown in Table 6.

     Tables 7 through 10 show correlation coefficients of regressions be-
tween selected water quality parameters and land use variables for all
subbasins in the Missouri Basin. Table 7 shows linear regression analy-
sis for pollutant mass flowrate (Ib/day) per square mile of watershed as
a function of each of the land use parameters per square mile of water-
shed. The regression equation is of the form Y = A + BX. Table 9 shows
results of regression analyses performed on data transformed into natural
logarithms of the same variables as in Table 7. The regression equation
is of the form UrCi = A + B AnX.

     Similar analyses were performed with water quality parameters (mgVliter)
and land use parameters per square mile of watershed both in normal  units
and in natural logarithms, and are shown in Tables 8 and 10.
                                   58

-------
                            Table 6.  POLLUTANT POTENTIAL OF VARIOUS FARM ANIMALS^/

Animal
Cattle
Hogs
Sheep
Chickens
BOD5
Ib/animal/day
1.0 - 1.5 (1.0)
0.2 - 0.56 (0.25)
0.05 - 0.15 (0.10)
0.015
SV
ratio—
1.0
0.25
0.10
0.015

0.26
0.032
0.02
0.003
Nitrogen
Ib/animal/day
- 0.49 (0.40)
- 0.05 (0.036)
- 0.03 (0.025)
- 0.0036 (0.0033)
N b/ P2°5
ratio—' Ib/animal/day
1.0
0.09
0.063
0.0067
0.12
0.025
0.012
0.0026
P2°5
ratio^/
1.0
0.21
0.10
0.022
N/P
15.30
6.58
9.56
5.80
a./  Data compiled from Loehrl/, Klausner et al.—' , and MRl£'.
b/  This ratio is standardized with respect to cattle (1.0). Thus, the BOD of waste from one chicken is 0.015
      times the BOD of waste produced by one beef animal.
I/  Loehr, R. C., Pollution Implications of Animal Wastes:  A Forward Oriented Review, FWPCA-WRC, Ada, Oklahoma
      (1968).
2J  Klausner, S. D., P. J. Zwerman, and T. W. Scott, Land Disposal of Manure in Relation to Water Quality, in
      Agricultural Wastes, Cornell Conference (1971).
3/  MRI, "The Pollution Potential of the Confined Livestock Feeding Industry," Final Report, EPA Contract No.
      68-01-0025, November 1971.

-------
                                  Table 7.
Pollutant Loading Versus Land Use:   Linear Regression Analysis
DEPENDENT
VARIABLE
Y
BOD-PPD/M2








P-PPD/M2







x
N03-PPD/M2








DS-TPD/M2








TUR/M2









DO . s






INDEPENDENT
VARIABLE
CAT/SO MI
HOGS/SO MI
LVSTK/SQMI
NITRO/SOMI
PHOS/SO MI
CROP/SO Ml
IRRI/SO MI
RAINFALL

CAT/SO MI
HOGS/SO MI
LVSTK/SQMI
NITRO/SQMI
PHI, , /SO MI
CROP/SO MI
II" . /SO MI
RAINFALL

CAT/SO MI
HOGS/SQ MI
LVSTh /SQMI
NITRO/SOMI
PHOS/SO MI
CROP/SO MI
IRRI/SO MI
RAINFALL

CAT/SQ-MI
HOGS/SQ MI
LVSTK/SOMI
NITRO/SOMI
PHOS/SO MI
CROP/SO MI
IRRI/SO Ml
RAIN) ALL

CAT/SQ MI
HOGS/SO MI
LVSTK/SOMI
NITRO/SOMI
PH^iS/SQ MI
CROP/SO MI
IRRI/S': Ml
RAINFAl
CAT/SO MI
HOGS/Si) MI
LVSTK/SOMI
NITRO/SllM
PHOi/SU MI
CROP/SO MI
I -Wl/SO Ml
RAINFALL
CORRELATION
COEFFICIENT
.1543
.1501
.1596
.4153
.2834
.3671
-.0568
.5634

.4254
.6729
.4547
.6384
.7430
.4599
-.2143
.6981

.2838
.3739
.2996
.101H
.2491
.2381
-.0361
.2799

.0951
.0825
.0946
.0140
.1387
.0599
-.0690
.1478

.4381
.2949
.4316
.2756
.3147
.2946
.0707
.030S
.1 . /
SIGNIFICANCE
LEVEL
N
.
.
.
4
•
.
*
•

*
.
.
*
^
•
•
•

,
,
•
•
.
•
•
•

m
.
•
.
•
.

•

•
.
•
*
.
*
•

192
198
184
007
052
016
361
000

200
072
182
066
045
179
342
061

000
000
000
121
002
003
347
001

133
167
134
435
052
242
222
041

000
003
000
004
002
003
269
368
.206
. 17f.SI .098
. 119H
.0601
.05
55
55
55
55
55
41
55
NO
1
1
1
. OF
OF
, 32
. 32
, 32
1, 32
1. 32
1. 32
1, 29
1. 32










1
1
1
1
1
1
1
1










1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1,

4
4
4
4
4
4
4
4

132
132
132
132
132
132
119
132

137
137
137
137
137
137
123
137

87
87
87
87
83
87
76
87
5J
53
53
53
53
53
3Si
53
                                           MEAN
                                          40.091
                                                  DBS.
                                                    34
62.869
34.211
73.929
7.084
2.hVH
.410
.039
22.246
12.13h
62.869
34.211
69.702
7.0B4
2.69H
.410
.03')
22.24fr
23.221
62.869
34.211
69.70?
7.084
2.69H
.410
.039
22.246
1.113
62.M64
34.211
66. 9M
7.084
2.69H
.410
.039
22.24*
.5B?
62.M69
34.211
66.951
7.0t)4
2.*9M
.4111
.U31*
22.246
62.«bV
34.211
66.VS1
7.0b4
2.b9H
.4111
.0.14
22.f-<."
233
233
233
233
228
233
201
233
6
233
233
?33
233
228
233
201
233
134
233
233
233
233
228
233
201
233
139
233
?33
233
233
228
233
201
233
89
233
233
233
233
228
233
201
233
233
233
233
233
22B
'13
201
233
                                                         STANDARD
                                                         DEVIATION
                                                            78.253

                                                            46.413
                                                            52.358
                                                            58.537
                                                            9.157
                                                            2.967
                                                             .289
                                                             .053
                                                            9.170

                                                            21.279

                                                            46.413
                                                            52.358
                                                            54.305
                                                            9.157
                                                            2.967
                                                             .289
                                                             .053
                                                            9.170

                                                           81.275

                                                           46.413
                                                           52.358
                                                           54.305
                                                            9.157
                                                            2.967
                                                             .289
                                                             .053
                                                            9.170

                                                            2.494

                                                           46.413
                                                           52.358
                                                           50.699
                                                            9.157
                                                             .967
                                                             .289
                                                             .053
                                                            9.170

                                                            2.443

                                                           46.413
                                                           52.358
                                                           50.699
                                                            9.157
                                                            2.967
                                                             .289
                                                             .053
                                                            9.170

                                                            ••6.413
                                                            5?.358
                                                            5:1.690
                                                            9.157
                                                            ?.967

                                                             .053
                                                            0.170
5PESSI1N FOUATION
Y = A + BX
A H
23.7333M .26018
32.41400 .22440
24.32204 .21330
14.94B ' 3.54923
19.9P412 7.47472
-.62941 99_,2JbO.L
43.37555 -83.83153
-66.86319 4.80790
-.12523 .19503
2.78125 .27345
-.28166 .17816
1.62649 1.4A360
-2.2391R 5.32819
-1.73322 33.81388
15.50303 -85.92784
-23.89911 1.61990
-8.01842 .49690
3.36754 .58033
-8.02820 .44833
16.81783 .90392
4.81480 6.82221
-4.20174 66.65695
25.38991 -55.35155
-31.95871 2.48050
.79129 .00511
.97819 .00393
.80112 .00465
1.08565 .00382
.79823 .11656
.90103 .51610
1.23971 -3.24113
.21862 .04019
-.86762 .02306
.11159 .01376
-.B0994 .02080
.061S3 .07353
-.11663 .25911
-.437*2 2.48741
.454HJ 3.25731
.40168 .00813
58.15171 .4H348
65.54765 .67231
57.05492 .47030
79.29812 1.30574
HO.JlflU^ 3 • OSO 1^
-.1.3055" 11 .17650
100. H3J84-31 J. 57171
34.61390 2.4P440
REGRESSION
SIGNIFICANCE
F
.78064
.73802
.83595
6.67038
2.79544
A*.M523
.09403
14.88301
.8838)
3.30905
1.04244
2.75238
4.93043
1.07281
.19253
3.60282
11.55900
21.44635
13.01260
1.36340
8.73098
7.92966
. 155*"1
11.2182
1.2514'
.93995
1.23728
.02698
2.66638
.49303
.58770
3.05860
20.66560
8.26802
10.90686
'.14991
9.12137
8.26824
.38225
.08099
.68241
1.71141
.77204
.19197
. 10985
1 .52934
.27453
.66975

-------
                                                          Table 8.
                           Water Quality Versus Land  Use:   Linear  Regression Analysis
DEPENDENT
VARIARLE

    Y

DO-MGL
 BOD-MGL
 P-HGL
I N03-MGL
 OS-TPO
 TIWHIOITY
              INDEPENDENT
               VARIABLE
CATTLE
HOGS
COWS*HOGS
LIVESTOCK
NITROGEN
PHOSPHATt
FERTILIZE*
IRRIGATED
.0051
.0491
.0305
-.0511
.0580
-.0325
.0554
.0055
              CATTLE
              HOGS
              CO*S*HOGS
              LIVESTOCK
              NITROGEN
              PHOSPHATE
              FERTILIZER
              IRRIGATED
              CATTLE
              HOGS
              COWS*HOGS
              LIVESTOCK
              NITROGEN
              PHOSPHATF
              FFRTILIZEH
              IRRIGATED
              CATTLE.
              HOGS
              COWS»HOGS
              LIVESTOCK
              NITR06FN
              PHOSPHATE
              FERTILIZE*
              IRRIGATED
              CATTLE
              HOGS
              COMS'HOtiS
              LIVESTOCK
              NITROGEN
              PHOSPHATE
              FERTILIZE"
              IRRIGATED
PHOSPHATI-
HAINFALL
C1TTLI
HUGS
f' (IPLAHO
 COKRFLATION  SIGNIFICANCE
.COEFFICIENT    ,  LEVEL
                                            .480
                                            .315
                                            .383
                                            .308
                                            .285
                                            .378
                                            .294
                                            .481
                              .031
                              .222
                              .098
                              .000
                              .016
                              .111
                              .023
                              .253
                              .104
                              .420
                              .220
                              .2B2
                              .270
                              ,3b8
                              .287
                              .319
                              .000
                              .000
                              .000
                              .000
                              .000
                              .000
                              .000
                              .000
                              .341
                              .167
                              .234
                              .?70
                              ,43ft
                              .110
                              .332
                              .170
                                            .019
                                            .-.?«
                                            .000
                                            .010
                                            .01S
                              .2314
                              .0960
                              .1615
                              .3991
                              .2646
                              .1589
                              .2469
                             -.0877
                              .3733
                              .0624
                              .2344
                              . 176s.
                              .1077
                              .1116
                              .1721
                              .1823
                              .4136
                              .2936
                              .3672
                              .2925
                              .4085
                              .3290
                              .4060
                              .3445
                              .0350
                              .0827
                              .11620
                              .0523
                              .0138
                              .1047
                              .0371
                              .0861
                                         NO. OF
                                         tY) PAIUS
                                            98
                                            98
                                            9A
                                            98
                                            98
                                            94
                                            98
                                            79
                                66
                                66
                                66
                                66
                                66
                                61
                                66
                                60
                                13
                                13
                                13
                                13
                                13
                                13
                                13
                                 9
                               135
                               135
                               135
                               135
                               135
                               135
                               135
                               121
                               139
                               139
                               139
                               139
                               139
                               139
                               139
                               12b
                               .n]9b
                               . (4HS
                              - . l> 1 H 0
                                                          M9
                                                          «9
                                                          7b

MEAN

8.669
62.B69
34.211
97.QBO
160.073
7.084
2.698
9.724
.039
6.819
62.869
34.211
97.080
160.073
7.084
2.698
9.724
.039
1.169
62.869
34.211
97.080
160.073
7. 084
2.698
9.724
.039
2.672
62.H69
34.211
97.080
160.073
7.084
2.698
9.724
.039
1603.262
62.B69
34.211
97.080
160.073
7.084
2.698
9.724
.039
327.090
2.6
-------
                                                      Table  7.
                                                           Table 9.
                      Pollutant  Loading Versus Land  Use:  Logarithmic  Regression Analysis
DEPENDENT
VARIABLE
BOO-PPU/M2
P-PPD/M2
 DS-TPD/M2
 TUR/M,
INDEPENDENT   CORRELATION  SIGNIFICANCE     NO. OF
 VARIABLE    COEFFICIENT      LEVEL

     X                                N      OF
             CAT/SO MI
             HOGS/SO MI
             LVSTK/SOMI
             NITHO/SQMI
             PHOS/SQ MI
             CHO.'/SQ MI
             IRRI/SQ MI
             RAINFALL
             CAT/SQ MI
             HOGS/SO MI
             LVSTK/SQMI
             NITRO/SQhl
             PHOS/SQ MI
             CROP/SO MI
             IRRI/SQ MI
             RAINFALL
 N03-PPD/M2
 00-PPD/M2
             CAT/SQ MI
             HOGS/SQ MI
             LVSTK/SUMI
             NITKO/SUMI
             PHOS/SQ MI
             CROP/SO MI
             IRR1/SO MI
             RAINFALL
             CAT/SU Ml
             HOGS/SQ Ml
             LVSTK/SQMI
             NITRO/SbMl
             PHOS/SO MI
             CROP/SU Ml
             IRRI/' '1 Ml
             UAINI   L
 T4T/SU  MI
 HOGS/SO "1
 LVSTlWbOMl
 NITRO'SUMJ
 PHOi/S   M|
 CKOP/bU MI
 JHM/SIJ .11
 HAINF Al L
                             .3007
                             .3860
                             .3313
                             .55H1
                             .5651
                             .3332
                             -.0881
                             .4158
                 .552*
                 .*912
                 .5542
                 .5551
                 .5971
                 .3708
                -.1863
                 .5401
CAT/SQ Ml
HOGS/SQ MI
LVSTK/SQMI
NITRO/SUM
PHOS/SQ MI
CROP/SO Ml
IRRI/SU MI
RAINFALL
.5063
.4876
.5165
.51H5
.53B6
.3610
.1191
.44*4
                 .1231
                 .1057
                 .1295
                 .0950
                 .1962
                 .02«7
                 .0482
                 .0729
                 .218
                 .£•455

                 .1968
                 .1824
                 .1877
                 .0352
                 .1815
                              ,1730
                              ,  --04
                              .2/66
                               1 IS?
.042
.012
.028
.000
.000
.027
.319
.007
.128
.161
.127
.126
.105
.235
.362
.134
.000
.000
.000
.000
.000
.000
.097
.000
.074
.108
.064
.133
.010
.369
.297
.1V7
.020
.010
.017
. 1) 1?
.047
.039
.3HO
.044
. l')3
.OH?
.OSU
.020
.002
.262
.295
.201
34
34
34-
34
34
34
31
34
6
6
6
6
6
6
6
6
13*
13*
13*
134
134
134
121
134
139
13V
139
139
139
139
125
139
89
89
89
89
85
89
78
H9
55
55
55
55
55
55
41
55
1, 32
1, 32
1, 32
1. 32
1, 32
I, 32
1, 29
1. J2
1, 4
1, 4
1. 4
1. 4
1. 4
1. *
1, *
1. *
1.132
1.132
1.132
1,132
1.132
1.132
1.119
1,132
1,137
1,137
1,137
1.137
1,137
1.137
1.123
1.137
1 87
1 87
1 87
1 87
1 83
1 87
1 76
1 . '7
1 53
1 V>J
1 53
1 53
1 53
1 53
1 39
1 53
                                                                   Mt'AN
                                                                   2.306
       STANDARD    REGRESSION EQUATION    REGRESSION
OHS.   DEVIATION   LN( Y)=AfB*LN(X)  SIGNIFICANCE

                       A         B           F
                                                                              34
3. 853
1.967
3.997
.U74
.102
•1.370
•4.028
3.017
1.116
3.853
1 .9f>7
3.93H
.874
.11)2
•1.370
•4.U2H
3.017
.747
3.M53
1.V67
3.93K
.874
.102
•1.370
•4.028
3.017
•1.177
3.«53
1 .Vf>7
3.V11
,Hlt
.102
•1.370
•4.02H
3.017
•2.72ft
3.HS3
1.967
3.911
.Uf4
.ID?
•1.370
•4.02*
3.017
J.I I1'
3.HS4
i . *o r
J . V 1 1
.H/4
.111?
•l.yi'
J . ii 1 '
233
233
233
233
228
233
201
233
6
233
233
233
233
228
233
201
233
134
233
233
233
? '3
228
233
201
233
139
233
233
233
233
228
233
201
233
89
233
233
233
233
228
233
201
?33
55
233
?33
233
233
2?B
233
?3J
                                                                                      1.883
                                                                                      1.P57

                                                                                       .805
                                                                                      2.223
                                                                                       .821
                                                                                      1.731
                                                                                      1.6*3
                                                                                      1.186
                                                                                      1.419
                                                                                       .417

                                                                                      2.459

                                                                                       .805
                                                                                      2.223
                                                                                       .821
                                                                                      1.731
                                                                                      1.6*3
                                                                                      1.186
                                                                                      1.419
                                                                                       .417

                                                                                      1.702

                                                                                       .805
                                                                                      2.223
                                                                                       .80*
                                                                                      1.731
                                                                                      1.643
                                                                                      1.186
                                                                                      1.419
                                                                                       .417

                                                                                      2.015

                                                                                       .805
                                                                                      2.223
                                                                                       .804
                                                                                      1.731
                                                                                      1.6*3
                                                                                      1.186
                                                                                      1.419
                                                                                       .417

                                                                                      1.821

                                                                                       ,B05
                                                                                      2.223
                                                                                       .804
                                                                                      1.731
                                                                                      1.64J
                                                                                      1.1 H6
                                                                                      1.414
                                                                                       .417
-.*0576
1.66265
-.75971
1.775*1
2.23964
3.03027
1.83*79
-3.355*6
-3.79714
.30840
-3.82171
.59532
1.0467H
1.91096
.11135
-6.13H24
-5.21*33
-.31357
-5.34*78
.10389
.66509
1.77265
1.57903
-7.15*68
-2.17981
-1.33610
-2.2*882
-1.2585*
-1.19770
-1.23335
-l.*096*
-2.07350
-*.R321*
-3.163*0
-*. 92006
-2.92587
-2.74858
-2.28900
-2.52456
-5.371U?
1.60612
? . 90740
. 4 Vo4b
d. 86057
3.07231
3.?9«J
-------
                                                     Table 10.
                         Water Quality Versus Land Use:   Logarithmic Regression Analysis
to
DEPENDENT
VARIABLE
LN(V)
DO-MGL








BOD-HGL








P-MGL








N03-MGL








OS-THO








TUPHIDITY






INDEPENDENT
VARIABLE
LMX)

CATTLE
HOGS
COWS«HOGS
LIVESTOCK
NIIROGEN
PHOSPHATE
FERTILIZE*
IRRIGATED

CATTLE
HOGS
COKS»HOGS
' TVESTOCK
NITROGEN
PHOSPHATF
FERTIL12FH
IRRIGATED

CATTLt
HOGS
COWS»HOGS
LIVESTDCK
NITROGEN
PHOSPHATF.
FERTILIZE*
IRRIGATED

CATTLE
HOGS
COWS'HOGS
LIVESTOCK
NITRObF.N
PHOSPHATE
FERTILIZE*
IRRIGATED

CATTLE
HOGS
COWS«HOGS
LIVESTOCK
NITROGEN
PHOSPHATh
FEHTILI /K«
IRRIGATED

PHOSPHAth
»AIi.. . LL
i'»T II I
MM'iS
CROPLAND
[•*nj(iATt i>
CORRELATION
COEFFICIENT


,06«9
.0826
.0710
-.02H1
.0117
-.0406
.0397
-.0592

.3«25
.4118
,')94fl
.5224
.325?
.2771
. 300fl
-.103H

.3689
.J123
.3549
.292?
.1427
-.0535
. M
-.f O4H

.569?
.5204
.5604
.S316
.5363
.4456
.5124
.1245

-.(1124
-.0053
.0019
.0744
-.0378
.0914
-.0007
-.0905

.i./63
.H99«>
. 1 il4'/'
.1/34
. ll?*!?
-. 141S
•• 'GNIFICANCE
LEVEL


.250
.209
.?44
.392
.454
.349
.349
.30?

.001
.000
.001
.000
.004
.015
.007
.215

.107
.149
.117
.166
.321
.431
.406
.246

.000
.000
.000
.000
.000
.000
.000
.OHH

.441
.47S
.491
.192
.329
.142
,4-'7
.15rt

.244
.177
.16"!
,«'->i'
.121
.10M
NO. OF
(X«Y) PA1
N

98
98
98
9H
98
94
98
79

66
66
66
66
66
61
66
nO

13
13
13
13
13
13
13
9

134
134
134
1 .14
134
134
134
l.-O

139
139
111
139
139
139
139
125

H5
H9
MV
rt9
t<9
It*
                                                           MtAN
                                                           2.134
                                                                  NO. OF
                                                                   OHS.
                                                                    9fl
3.H53
1.967
4.11^
4.634
.874
.!.!/•
1.2B3
-4.02H
1 . 464
3.HS3
1.967
4.11,-
4.6 ^
• fl74
.1(1?
!./•«'<
-4.0?!-
-1. !<•>.
3.hbl
1 .96/
4.11?
4.6J4
.874
.10?
1.2H f
-4.0?h
.?94
3. (.51
1.9r,7
4.11?
4.6J^
.H74
.10?
1.2"t
-'i.Of'H
S . n U
3. MSI
1 .96 /
4.1
4.6J-*
.874
.10?
i.2
-------
     The independent variables (X) and dependent variables (Y) of Tables
7 through 10 are defined as follows:

                         Y, Dependent Variable

     BOD 	 Biochemical Oxygen Demand, Ib/mile2/day
     P   	Phosphorus               , Ib/mile /day
     N03 	 Nitrate                  , Ib/mile2/day
     DS  	 Dissolved Solids         , tons/mile2/day
     Tur 	 Turbidity                , Jackson turbidity units/mile2
     DO  	Dissolved Oxygen         , Ib/mile /day
                        X, Independent Variable

     CAT      	number of cattle/mile2
     HOGS     	number of hogs/mile2
     LVSTK    	number of livestock (as cattle)/mile2
     NITRO    	 nitrogen fertilizer applied, tons/mile /year
     PHOS     	 phosphate fertilizer applied, tons/mile /year
     CROP     	cropland, mile /mile  of basin
     IRRI     	irrigated land, mile2/mile  of basin
     RAINFALL 	 rainfall amount, in/year
     A comparison of the  regressions is made in terms of a statistical
test (F-distribution)  for the goodness of fit to assess the usefulness
of the data base. Table 7 which  is based on pollutant loading (Ib/mile2
/day) shows 14  statistically significant linear regressions at 95% con-
fidence  level based on F-distribution. Table 8 which is based on concen-
tration  (mg/liter) of  water quality parameters, also shows 14 significant
linear regressions at  the same level. The number of significant regres-
sions in Table  9 and Table 10, which are based on logarithms of the same
variables  as in Tables 7  and 8,  are, respectively, 17 and 14. Nitrate,
BOD, and turbidity yielded the largest number of significant regressions.
The results of  Tables  7 through  10 may be summarized for the number of
significant regressions for each water quality parameter as follows:
                    SUMMARY OF SIGNIFICANT REGRESSIONS

Table No.      BOD      P      Mh     D§     DO     Turbidity     Total
     7           30600          5           14
     8           20800          4           14
     9           40712          3           17
    10-          70700          0           14
                                    64

-------
     There appears to be very little difference in total number of regres-
sions when linear regressions are compared with logarithmic regressions.
BOD, DO, and DS yielded more number of significant regressions while turbi-
dity yielded less on a logarithmic scale. It appears that,  in general,
both linear and logarithmic regressions can yield equally significant re-
gression equations to describe adequately the given data.

     Figures 29 through 34 describe graphically some of the regression
analyses shown in Tables 7 through 10.  These plots are summarized as fol-
lows.
Figure No.                    Regression Equation

    29          BOD, mg/liter = 0.88 (cattle/mile2)0'39
    30          BOD, Ib/mile2/day = 0.47 (livestock/mile2)0'77
    31          BOD, Ib/mile2/day = -0.63 + 99.28 (cropland/mile2)
    32          N03, Ib/mile2/day = 0.006 (cattle/mile2)1-55
    33          Turbidity, JU = -75.5 + 6.4 (cattle/mile2)
    34          N03, Ib/mile2/day = 0.90 (N applied/mile2)0-74
     These regressions are significant at the 95% confidence level.

Multiple Regression Analysis

     Critical examination of the results of bivariate regression analyses
was made to determine significant functional relationships between water
quality parameters and land use parameters. This examination revealed that
these analyses could be used to determine the selection of independent
variables to be subjected to stepwise multiple regression analyses.

     Four water quality variables were chosen for the analysis as inde-
pendent variables. These were:  (a) turbidity (JU);  (b) turbidity (JU)
x flow rate (cu ft/sec)/mi2; (c) BOD (Ib/mile2/day);  and (d) nitrate
(Ib/mile2/day). The results of bivariate regression analyses as shown in
Tables 7 through 10 and summarized on page 64 show that the total number
of significant regressions for BOD, N03, and turbidity was 16, 28, and
12 respectively. Other parameters had much fewer number of significant
regressions. Therefore the multiple regression analysis was conducted
for BOD, N03, and turbidity. Since turbidity was measured in Jackson
turbidity units (JU) as concentration of suspended matter, it was de-
cided arbitrarily to convert the concentration units to mass units by
multiplying with the flow rate. However, no attempt was made to express
these units in Ib mass flow rates because of a lack of correlation be-
tween turbidity (JU) and mass concentration of suspended solids (mg/A).
                                   65

-------
   5r
~
8
co
                                                                                       In Y = -0.13+0.39(lnX)
   -1
                           234567

                                                    CATTLE/MI2, In X
                      Figure 29.  Regression of BOD-MGL versus cattle  per square mile
8
10

-------
  >-
  c
  "   2
'  Q
  a.
  a.
  i

  O

  O
  CQ
ON
~J
                                                                 In Y=-0.76+0.77(In X)



                                                                 F=3.95



                                                                 F0>05=4.17



                                                                 n=34
     -2
     -41
                                                                       J_
                 Figure 30.
2345678



                         LIVESTOCK/MI2, In X



Regression of BOD-PPD per square mile versus  livestock per square mile
                                                                                                                    10

-------
           280
00
           240
           200
           160
        a.


        Q

        O 120
        CO
            80
            40
          Y=-0.63+99.28(X)



          F=4.99




          F0.05 = 4-17



          n=34
                                             i	1.1   . i	I	I	I     I     I 	I	i     I  •  i     i     j
                       0.1
                   Figure 31.
0.7
0.8
                      0.4        0.5       0.6


                          CROPLAND/MI2, X



Regression of BOD-PPD per square mile versus cropland  per square mile
0.9
1.0

-------
          8r
      -£  2

      CM*




      I
      O.
                                                                                          In Y=-5.21 +1.55(ln X)



                                                                                          F=45.49







                                                                                          n = 134
\o
         -2
         -4
-6
                    J	L
                                              J	\
J	L
                                                                                            J.
                     1          23456789


                                                       CATTLE/MI2.  In X


                    Figure  32.   Regression  of N03-PPD per  square mile versus  cattle per square mile
                                                                                                       10

-------
  lOOOr
   800
   600
>:  4oo
D£
   200
                                                                     Y = -75.50+6.40(X)
                                                                     F = 10.51

                                                                     F0.05 = 3-90
                                                                     n = 77
  -200
j	i	 j   	i	i	i	i	i	i	i	i	i	i	    i
                                                                        i	i
                10
20
•30
40         50
     CATTLE/MI2, X
60
70
80
90
100
                 Figure 33.  Regression of turbidity,  JU versus  cattle per square  mile

-------
-3
-2
-1
0         1          2
   NITROGEN/MI2, In X
                                                                      In Y = 0.11 +0.74(ln X)
                                                                      F =48.53
                                                                           = 3.92
Figure 34.   Regression of N03-PPD per  square mile versus nitrogen per square mile

-------
 Thus turbidity x flow rate was used as a dependent variable in the re-
 gression analysis.

      The eight analyses were carried out for multiple regression  on the
 following four dependent variables:  (1) turbidity (JU)  x flow (ft3/sec)
 per square mile,  (2) BOD (Ib/day) per square mile, (3) nitrate (Ib/day)
 per square mile,  and (4) turbidity (JU) per square mile.  Both  natural
 logarithmic transformations of data as well as the normal data were used
 in the regressions.
The regression equations are of the general form

                            n
                       i +  2  Ai -i * xi i
                            =   X»J    1>J
                       Yi = c
 where C = Constant

       A = Coefficient

       X = Independent variable

       Y = Dependent variable
     Ten independent variables were chosen for analysis. These included:
(1) cropland in mile2/mile2 of the basin area; (2) pasture land in
mile2/mile2; (3) irrigated cropland in mile2/mile2; (4) number of cattle
/mile2; (5) number of hogs/mile2; (6) nitrogen fertilizer in tons/mile2;
(7) phosphate fertilizer in tons/mile2; (8) rainfall amount in in/year;
(9) slope percent class; and (10) cover factor.

      The values of cover factor and livestock were evaluated  on  a weighted
 basis, consistent with values reported in the literature.  The cover  factor
 was computed by dividing all the land in a watershed  into  four groups (viz.,
 row crops,  small grain, pasture and hay,  and woodland), and assigning a
 weight to each group.  Thus,  row crops were weighted 0.45,  small  grain 0.2,
 pasture,  0.02,  and wooded land 0.005.—  The sum of. all the four  weighted
 \J  Great Lakes Basin Framework Study, Appendix No.  18,  "Erosion and  Sedi-
       mentation," Draft No. 3, Work Group on Erosion and Sedimentation,
       September 1971, pp.  18-33 (for row crops, cover factor was modified
       to reflect adequate management conditions).
                                    72

-------
groups is divided by the area of the watershed,  resulting in the weighted
cover factor.

     Table 11 shows a summary of the results of  the multiple regression
analyses, including the coefficient of variation for each regression
equation.

     Each of these equations describes a correlation between a dependent
water quality parameter and several independent  parameters which include
land use characteristics, physiographic characteristics and other related
parameters which affect water quality in the streams of the Missouri Basin.
These correlations are not, however, strictly valid beyond the conditions
existing when regression data were collected.

     The complexity of the regression equation increases if either the
number of independent parameters increases, or true "independence" among
these parameters does not exist. Both these conditions must be recognized
in these analyses in properly interpreting the significance of the regres-
sions.

     BOD loading is represented in Table 11 by an equation containing six
independent parameters, and may be written as follows:

     BOD, Ib/mile2/day = -24.3 - 374.5 (cropland, mile2/mile2)

     +360.0 (irrigated land, mile2/mile2) - 0.4  (cattle/mile2)
                    2
     -1.7 (hogs/mile ) + 5.4 (rainfall, in/year) + 2,831 (cover factor).

     Similarly, nitrate loading is related to seven independent parameters
as shown by the equation,

     N03, Ib/mile2/day =

     -1.3 + 114.7 (cropland, mile2/mile2)

     -322.5 (irrigated land, mile2/mile2)

     +0.3 (cattle/mile2) - 0.4 (hogs/mile2) +

     1.1 (nitrogen applied, tons/mile2/year) +

     1.0 (rainfall, in/year) + 578.1 (cover factor).
                                    73

-------
Table 11.  SUMMARY OF MULTIPLE REGRESSION ANALYSES
General equation: Y(I) = C(I) + E A(I,J) * X(I,J)
Dependent variables Independent variables
n
X(l) = Cropland/mile 2 Y(l) = Observed value of turbidity (JU) * flow rate (cfs)
X(2) = Pasture land/mile 2 Y(2) = Observed value of BOD Ib/mile2/day
X(3) = Irrigated la.nd/mile Y(3) = Observed value of N03 Ib/mile2/day
X(4) = Cattle/mile Y(4) = Observed value of turbidity (JU)
X(5) = Hogs/mile2 2 G = Constant
X(6) = Nitrogen/mile A = Regression coefficient
X(7) = Phosphate/mile
X(8) = Rainfall
X(9) = Slope-Percent-Class If
X(10) = Cover Factor Degrees significant
of at 95%
yd) c(i)
Y! -561.0
Y2 -24,3
Y3 - -1.3
Y4 -0.1
Mill ^
InY^ -8.6
lnY2 -0.8
lnY3 -11.2
lnY4 -8-9
4,905
-374
114
3
IBX.1
-4
-2
-1
-2
.5 -
.5
.7 -
.-3 -
InX?
.5 -
.3 -
.9 -0.2
.0 -
6.6 -10.
360.0 -0.4 -1.
-322.5 0.3 -0.
-
2 -
7 . »
4 1.1
0.1
InX1^ InX^ lnX5 InX^ InX*?
2.5 0.
-0.03 -0.03 -0.
-0.1 1.8 -0.
1.5 -0.
2 - 0.7
2 -
2 0.6
2 - -0.5
21.5 510.6 23,098
5.4 - 2,831
1.0 - 578.1
0.4 10.4
^•n^8 InXq InX^Q
-1.1 1.3 0.4
2.6 - 1.7
1.2 - 0.6
0.9 0.5 1.5
freedom
DF
6,
6,
7,
7,
DF
7,
6,
8,
7,
40
24
113
77

10
24
112
73
Statistic confidence Value of
F * Y(I)
1
6
7
4

0
1
7
2
.42 *
.34 *
.11 *
.57 *
F *
.51
.13
.55 *
.54
342.9
34.9
13.6
0.6
InY(I)
-2.81
3.29
2.24
0.61




mi
0.06
26.84
9.34
1.84

-------
     Both these equations were shown to be significant at 5% level.
                                                               o
Using these equations, it was calculated that 34.9 Ib BOD5/mile /day and
13.6 Ib N03/mile /day are being discharged into the Missouri River Basin.

     Similar calculations using natural logarithms yield 26.84 Ib BOD/mile2
/day and 9.34 Ib NC^/mile^/day. However, the regression equation for BOD
based on logarithms was shown to be not significant at 5% level.

     A review of Table 11 indicates that the usefulness of multiple  regres-
sion equations is presently limited to the current data base and the regions
under study. The following are comments that can be made of Table 11.

     1.  These multiple regressions are the results of analyses relating
selected nonpoint source variables to water quality parameters, based on
massive historical data. These regressions show that correlations do exist
among independent variables and dependent variables. However, the results
are still preliminary and have limited value for use in making prediction
of water quality. Particularly, these equations should not be used to pre-
dict water quality outside of the Missouri River Basin where a different
structure of relationship might be expected.

     2.  Careful scrutiny is required for all variables that are included
in the multiple regression equations even though they were found to be
significant in a bivariate regression analysis. Some regression coefficients
are not physically justified with regard to their sign and/or magnititude.
This might have been caused by the close interrelationship often called
"multicollinearity" among the "independent variables". Multicollinearity
may not be an obstacle in itself when a regression equation is to be used
for prediction; but it can be a serious problem when one wishes to draw
conclusions about the causal effects of specific independent variables.

     3.  The nonpoint source factors are related to the water quality
parameters through some variables which were not included in the regres-
sions, particularly those variables which relate to pollutant generation
and to pollutant transport. These variables were necessarily omitted be-
cause adequate data do not exist, or because factors have not been iden-
tified. As a result, these regression equations should not be used directly
to interpret "cause-and-effect" relationships, nor should they be utilized
in attempts to control water pollution by changing the values of the few
independent variables included in these regressions.

     Consequently, we feel strongly that further collection and analysis
of specific data are warranted. Recommendations for specific actions are
made in Section VI.
                                    75

-------
                               SECTION VI

                    CONCLUSIONS AND RECOMMENDATIONS

     The components of a generalized nonpoint pollution model are:   (1)
natural source characteristics—rainfall, soil characteristics,  slopes-
-which help to determine what the land is used for, and also define in-
herent tendencies to generate or release pollutants; (2) man-induced source
characteristics—essentially the land use pattern, including inputs of
potential pollutants such as pesticides and fertilizers; (3) properties
of the pollutant (water solubility, absorption characteristics,  lifetimes,
chemical transformations); (4) pollutant generation and transport mecha-
nisms and rates, which are functions of (1), (2), and (3);  and (5)  water
quality data which reflect pollutant inputs from both nonpoint and  point
sources. The Data Bank generated in Phase I and evaluated in Phase  II(A)
contains extensive information on Items 1, 2, and 5, i.e.,  natural  and
man-induced characteristics of the land, and the quality of water in the
Missouri River Basin. It contains essentially no information on the proper-
ties of potential pollutants and on the mechanisms involved in transport
of these pollutants from the land to surface waters. This program to date
has consisted of an evaluation of the possibility that useful nonpoint
models might be constructed by a systematic analysis of a voluminous body
of data which exclude Item 3, pollutant properties, and Item 4,  mechanisms
and rates of pollutant transport. Model development did not exclude the
use of existing information on pollutant properties and transport mechanisms.
However, information in these areas was generally inadequate.

     The program also involved tests of the adequacy or completeness of
data on land and water characteristics relevant to model development. In
this regard, three questions have been asked:  (1) are data on parameters
included in the Data Bank accurate or reliable?; (2) are individual param-
eters reported for a high percentage of the stations, so that coverage
is adequate?; and (3) are additional parameters useful or necessary?

     The sum total of the correlations, plots, and subbasin-by-subbasin
analysis presented in Section IV quite obviously comprises an overall model
of present (1969) water quality in relation to nonpoint sources of  pollu-
tion. This model has limited predictive capabilities, and efforts to ex-
tract equations which quantitatively relate specific effects to causes
                                   76

-------
met with partial success.  Several conclusions  developed  from the  Phase
II(A) study which relates to the above questions  and to  the  structuring
of further study are presented below,  followed by an assessment of  the
overall significance of the study relative to  advancing  a  capability  to
quantify nonpoint pollution, assess its impact, and  develop  effective over-
all control measures.

CONCLUSIONS

     Specific conclusions drawn from Phase II(A)  are as  follows:

A General Model

     The data base very clearly indicates that increasingly  intense analy-
sis of agricultural operations is accompanied  by  increasing  water pollution.
This fact is apparent from inspection of data, and is confirmed by  compu-
terized mathematical analysis. The significance of this  is considered to
be greater than a mere confirmation of accepted knowledge  and theory, for
it has been obtained for a large, real world system described by  a  massive
body of data. It confirms research plot and research watershed results, on
a greatly expanded scale. The present study also  has yielded apparent ex-
ceptions to expected correlations. Explanations of these apparent exceptions,
developed in further study, should prove to be an important  part  of the
body of knowledge of nonpoint pollution, particularly with regard to  interac-
tions between natural factors and man-induced  factors. In  further study,
time can profitably be spent on refinement of  the general  model developed
in Phase II(A), with emphasis on development of a rationale  for summing
up agricultural activities and relating the sum to water quality.

Source-Pollutant Models
     Analyses to date have yielded certain specific cause/effect  equations
which pass tests for validity over the entire basin,  but which do not ade-
quately describe selected subsets of data (subbasin data).  The equations
have a basic defect in that all of the load of a particular pollutant,
except sometimes for a constant baseline load, appears to come from only
one of a few of the total number of known sources.  It has been concluded
that the regression models developed in Phase II(A) permit one to state
with confidence that, supported by statistical tests, a basinwide correla-
tion between a water quality parameter and a source or sources exists;
the regression models are not, however, sufficiently well developed or
understood to permit their use in calculation of stream pollutant loads.
In our judgment, further development or analysis aimed at specific and
quantitative models, and based on a purely statistical approach,  will not
be fruitful with the present Data Base. This conclusion does not rule out
                                    77

-------
further statistical analysis aimed at strengthening tentative  conclusions,
and refining and expanding correlations.

Test Results Consistent with Accepted Theory

     The basinwide analysis of data by computerized methods yielded over-
all results which are consistent with accepted concepts and known facts
about nonpoint pollutants. Nitrate, BOD, and turbidity (suspended sedi-
ment) correlate with intensities of agricultural operations; these pol-
lutants should be related in a straightforward manner to runoff and the
density of tillage, livestock, fertilizer, etc. Certain other  pollutants,
notably phosphorus, dissolved solids, and dissolved oxygen did not cor-
relate with land use intensity. Phosphorus emissions are determined in
part by geochemistry, and phosphorus deposited in streams is heavily ad-
sorbed on deposited silt. Dissolved salts are influenced heavily by geo-
chemistry, and dissolved salt concentrations are notably higher in the
less heavily farmed regions of the basin. Dissolved oxygen levels are gen-
erally fairly high in the Missouri Basin; while DO responds broadly to
agricultural intensity, other factors such as natural reaeration are ap-
parently pronounced enough that DO is not a sensitive measure  of pollu-
tional inputs.

    One can justifiably be pleased that analyses of a very broad data base
confirm expectations based on known properties of pollutants and accepted
theories regarding generation and transport mechanisms. This aspect of
the results of data analyses (i.e., that correlations were found with
cause/effect parameters which should in theory correlate strongly, and
were not found where correlations might be expected to be weak) is strong
evidence that the Data Bank presents an adequate and honest picture of
land use and water quality in the Missouri Basin.

Gaps in Water Quality Data

     Water quality data for the 233 stations varies in completeness or
adequacy from parameter to parameter. Particularly noticeable is the scar-
city of available data on pesticides and BOD. Model development is not
feasible for pesticides. BOD is reported for only 68 of the 233 stations
(29%). With these two exceptions, coverage of water quality parameters
is adequate for the 233 stations.

Land Use Coverage

     Land use coverage is on the whole excellent. Coverage based on the
county unit has been calculated for the areas represented by each of the
233 stations, and can readily be compiled for other geophysical boundaries
as has been done for the 13 subbasins.

                                   78

-------
     Land use data in the Data Bank were acquired directly from available
data compilations, with two exceptions—namely fertilizer and pesticide
use. Data on these two parameters were obtained by questionnaire (pesti-
cides) and calculation from available information (fertilizer use).

     Partial updating of the Data Bank can be effected  annually if desired.
This process can be accomplished most easily every 5  years, however, by
computer accession to the Census of Agriculture.  Special  procedures would
be required in the updating process for collection or estimation of pesti-
cide and fertilizer use at the county level.

Coverage of Basic Land Properties

     Data on basic land or area properties are extensive, and of high  quality.
The potential usefulness of the Data Bank could,  however, be  substantially
increased by addition of certain information. Of particular interest are
the following:

     *  Precipitation mass balance data:  runoff, infiltration, and  evapo-
        transpiration.

     *  Soil property data:  the USDA is in the process of an extensive
        characterization of soils throughout the country in terms of basic
        soil properties, many of which relate to  water  quality and nonpoint
        pollution. In addition, soil losses (erosion) have been calculated
        for basic land uses throughout the United States; this information
        is a prime candidate for inclusion in the Data  Bank,  after it  has
        been studied to determine how it might best be  used.

Adequacy of Spatial Representation of the Basin

     The average watershed associated with each of the  233 stations  is
over 2,000 miles  in area. Each station represents the  quality of water
in about 300 miles of surface streams, and reflects pollutional inputs
from a large watershed which is extensively diversified in land use.

     In order to achieve the ultimate objective of development of stream-
source models, it is essential that the water quality stations uniformly
interact with their respective watersheds. For example, the load of  nitrate
or BOD at a station should either be the cumulative load deposited through-
out the watershed, or some uniform percentage of that load. With the pos-
sible exception of dissolved salts, none of the standard array of water
pollutants will be truly conservative over such long stretches of rivers,
and the "load" at a station will be a fraction of the deposited load.  With
233 stations representing large watersheds of differing sizes and basic
                                   79

-------
characteristics, one is not surprised to find that source-to-station water-
shed relationships are reasonably uniform for some pollutants, and are
not for others. The size of the watershed represented by a water quality
station appears, ^herefore, to be a critical factor, and the results of
this study indicate t1. , *• 2,000+ miles2 is on the verge of being too large.

     Success in allocating t 5tal loads of specific pollutants among several
sources (either multiple .;ources of the same type, or different types of
sources) is highly dependent on the nature of the station-watershed re-
lationship. A  large load deposited 30-50 miles upstream may, for example,
register the same as a small load deposited 1-5 miles upstream of the sta-
tion. It is self-evident that the specificity of source-station relation-
ships will improve as the size of the watershed is diminished. This study
had only limited success in efforts to distribute loads among several sources,
which leads to the conclusion that the watersheds are too large and contain
too diverse an array of sources to permit achievement of this objective.

     This limitation of the Data Bank and its applicability to nonpoint
modeling can be stated in another way, as follows. The water quality at
a  station in a surface stream is a rather complex function of the loca-
tions and quantities of discharge upstream of the station, and of physical/
chemical/biological processes in the stream. The larger the watershed,
and the more diverse it becomes, the greater is the uncertainty in pin-
pointing pollutant contributions from specific sources, even though the
water quality  and source characteristic data may be both complete and ac-
curate. The results of this study tend to indicate that the water quality
station/watershed combinations available for this study are sufficient
in number to permit development of satisfactory relationships depicting
general interactions between basic land use patterns and water quality,
but are usually insufficient in number to permit specific and quantita-
tive description of source-stream interactions.

     The above conclusion will no longer hold if the sources themselves
can be more completely described. For example, sediment losses from the
varying land uses can be predicted, and the incorporation of data or func-
tions for sediment losses in the Data Bank would permit a more accurate
and specific modeling of erosion and of sediment deposition in streams.
Similarly, a model for nitrogen cycle and transport within the environ-
ment could be  combined with land use data to calculate quantities of nitro-
gen discharged from specific sources and combinations of sources.

Overall Adequacy of the Data Bank

     Whether or not the Data Bank is considered to be adequate depends
on the intended use. If the desired use is description of land use, land
characteristics, and water quality in such a manner that broad stream-land
                                   80

-------
use relationships can be easily visualized,  and changes  in  these  relation-
ships followed over extended time periods,  the Data Bank can  be considered
to be very nearly adequate, and updating activities need little expansion
in scope or in data coverage. If, on the other hand,  further  model  develop-
ment is the objective, and if it is desired in addition  that  the  Bank be-
come a vital part of a long-term planning and management system which en-
compasses water quality, nonpoint pollution control,  and land and resource
management for the Missouri Basin, it is evident that the Bank is inadequate.
In this regard, consideration should be given to the following:

     a.  Selective addition of water quality stations. This process should
take place with the long-term future in mind, and include possible  restruc-
turing of the present system of water quality stations.

     b.  More complete coverage of BOD.

     c.  Inclusion of point source information.

     d.  Significantly upgraded coverage of pesticides in surface waters.

     e.  Expanded, more specific, data on pesticide and  fertilizer  use.

     f.  Inclusion of data on runoff and infiltration.

     g.  Inclusion of systematized data on soil losses.

     h.  Addition of basic geochemical data, and of basic data on soil
properties.

OVERALL ASSESSMENT

     The Phase II(A) program was initiated with the expectation that the
groundwork established in Phase I and Phase II(A) would  be  the basis for
continuing nonpoint modeling activities in a Phase II(B) program. This
expectation was tempered by a realization that the Phase II(A) study could
yield results and conclusions which might necessitate a  reexamination of
objectives and/or a reassessment of methods for achievement of objectives.
The investigators have, as indicated in the conclusions  presented earlier,
come to a three-part overall assessment.

     1.  The Missouri River Basin Data Bank is a quite useful collection
in that it affords the means to (a) depict clearly the basinwide  interde-
pendence between land use and water quality, (b) demonstrate by mathematical
techniques that certain water pollutants are related to  land use, and (c)
show, also by mathematical means, that other water pollutants are either
less simply related to land use, or that other factors obscure simple rela-
tionships. However, this Data Bank has basic limitations and inadequacies—

                                    81

-------
in its present state—which preclude detailed development of predictive
models, and will limit continuing study to further analyses of trends and
correlations. Further study of this type will be more dependent on scientific
bias and investigator input, and less dependent on statistical data analysis.

     2.  Information and data required for useful expansion and updating
is at hand, and model development can proceed in an orderly fashion if
the additions are made.

     3.  The Data Bank is a unique and excellent base for updating and
expansion  to serve  several purposes—water quality management planning,
nonpoint source control and agriculture resource management, and land use
analysis and planning. The utility of the Data Bank will be enhanced by
systematic addition of a variety of data which relates broadly to water
and resource management.

RECOMMENDATIONS

     Several options for continuing study have been considered. One option
is a Phase II(B) follow-on to the present program based upon the current
Data Bank  with significant additions, and incorporation of submodels such
as the  Soil Loss Model as an essential part of continuing model develop-
ment. A second option involves intensive study of one or two subbasins,
and would  include the use of a model such as EPA's Auto Qual.—' A third
option  consists of  development of a master plan for upgrading and main-
taining the Data Bank, including designation of additional water quality
stations and delineation of all data which should be accumulated and re-
ported  routinely. These options are elaborated upon below.

Continuing Basinwide Model Development

     As stated earlier, the Data Bank in its present form limits model
development to further analysis of trends and correlations. More of this
type of analysis can profitably be conducted, but the greatest pay-off
will come  from analysis of a Data Bank expanded in such a manner that it
includes information or data which relate to the mechanisms and rates of
generation and transport of specific pollutants. Of first priority is the
inclusion  of available data on erosive losses of soils.
 I/ "Auto-Qual Modelling  System," U.S. Environmental Protection Agency,
~~   Office  of Air and Water Programs, Washington, D.C. (1973) (Supple-
     ment  I  - Modification  for Nonpoint Source Loadings.)
                                   82

-------
     The U.S. Department of Agriculture has documented certain information
by Land Resource Area (LRA), and other information by county,  which together
will permit a systematic calculation of sheet/rill erosion throughout the
Missouri Basin as a function of land use. Data by LRA includes:

     *  Slope (S), slope length (L), and erodibility factor (K)  (factors
        in the Universal Soil Loss Equation)—  for soil capability classes
        (a total of 29 possible class/subclass designations).

     *  Calculated soil loss data for cropland in the soil capability
        classes, with various typical crop rotations, tillage  practices,
        and conservation practices.

The county data, presented in the Conservation Needs Inventory,  include:

     *  Acreages of crops, hay, pasture, and other land uses,  by soil cap-
        ability class.

     *  Acreages of cropland under adequate conservation practices.

     The data on K, S and L, in combination with R (rainfall factor) values
(available from USDA handbooks), yield the basic erodibility of land, i.e.,
RKSL, a factor which one multiplies by a cover factor C and a practice
factor P to calculate sheet/rill erosion. The cover factor C is known for
various types of crops, and the practice factor P is a known factor for
various conservation practices. It is within our means, therefore, to in-
clude in the Data Bank all the information needed to calculate sheet/rill
erosion losses throughout the Missouri Basin. An additional refinement
involves a factor termed the delivery ratio, which is that fraction of
soil lost from fields which actually reaches a stream bed. Delivery ratios
are a function of soil type and drainage area size; while less precisely
established than the factors of the Universal Soil Loss Equation, accepted
values are available. Delivered sediment to streams is therefore calculable,
with information presently at our disposal.

     The importance of sediment as a pollutant has been both under- and
over-emphasized. The investigators view sediment as a very significant
pollutant in its own right, but not necessarily the most important one,
2]  Wischmeier, W. H., and D. D. Smith, "Predicting Rainfall-Erosion Losses
      from Cropland East of the Rocky Mountains," Agriculture Handbook
      No. 282, ARS-USDA, May 1965.
                                   83

-------
except perhaps in situations where streams/reservoirs are flagrantly fouled
with sediment. Sediment is, on the other hand, perhaps the most important
of the nonpoint pollutants in that it is a carrier of other pollutants
(nutrients, pesticides, metals, organics, bacteria, etc.), and for the
additional reason that it serves as an indicator of other pollutants.  One
would expect, therefore, to find a strong correlation between nutrient
emissions and emissions of sediment.

     For these reasons, it is strongly recommended that specific soil loss
information be developed for inclusion in the Data Bank, and that the fol-
lowing analyses be  conducted:

     *  Calculation of sediment emissions to streams, by subbasin, and
        for the 233 watersheds represented by the Station Master File.

     *  Development of relationships between sediment emissions and stream
        turbidity and suspended solids loadings.

     *  Mathematical analyses of possible relationships and correlations
        between sediment emissions and other water quality parameters,
        such as nitrate, phosphorus, and coliforms. If statistical tests
        are favorable, define models which.relate sediment emission to
        emissions of as many other pollutants as justified by adequacy
        of data.

     *  Analysis of the relative contributions of such pollutants as nitro-
        gen and phosphorus from different sources such as cropland in row
        crops, cropland in small grains, and livestock.

     *  Delineation of projected sediment emissions which would be the
        result of implementation of assumed control procedures, and exten-
        sion of projections to other sediment-related pollutants.

     Continuing study aimed at model development should include three ad-
ditional activities.  The first activity consists of further study of ferti-
lizers and pesticides addressed jointly to improving the information in
the Data Bank on use patterns and to development of models which are source-
specific (i.e., associate emitted quantities with livestock populations
and cropping patterns).

     The second activity consists of a renewed, essentially continuing
examination of recent (about 1970 to the present) data entered into, and
available from various data compilations. Some of the newer data may be
useful for increasing the reliability of the 1969-based data now in the
Bank. Year-to-year  trends may perhaps be perceptible by visual examina-
tion.
                                    84

-------
     The third activity consists of the study of  other known  and  avail-
able data compilations, assessment of possible utility in  the Data  Bank,
and addition of such new types of data (other than  soil  loss  information
discussed above) as will improve the capability to  analyze and model non-
point pollution. Data or. runoff and i  iltration  are  potentially  useful,
as is any information which helps to   aluate natural background  levels
of pollutants.

Subbasin Modeling

     The entire Missouri River Basin is too  large to  be  a  suitable  working
tool for intensive study aimed at specific and precise model  development.
Within the basin there are, however, many macro-scale "watersheds"  (each
several hundred to a few thousand square miles in area)  which represent
differing rural environments and will serve as excellent tools  for  detailed
model development. In a well-planned study,  such  modeling  activities could
be complemented by evaluations of the efficacy of nonpoint pollution con-
trol processes.

     In a study limited to one or two of these subbasins or watersheds,
it will be feasible to update land use to the current condition;  to collect
samples and measure water quality at a preselected  number  of  stations  in
addition to existing stations; and to verify and  apply a water  quality
model to be used for projections and to assist in analyzing defects or
voids in the information base. The EPA "Auto Qual"  Model is a candidate
for use in such a study. The water quality model  would provide  the  means
to calculate what happens to nonpoint pollutants  (and point pollutants
as well) after they reach surface waters, so that watev  quality data can
be more accurately matched with data on pollutant inpucs.

     Such an undertaking should not be casually initiated, for  a  few sober-
ing uncertainties exist. Chief among these is the problem  of  dealing with
slug inputs of pollutants when much of the existing water  quality data
is not specifically related to slug inputs.  An extensively instrumented
and monitored micro watershed or research plot is a much more comfortable
laboratory. The macro watershed must be the eventual  testing  and  proving
ground for nonpoint models, however. In a large and relatively  complex
area it will be possible to begin to identify and quantify interactions
between sources and/or pollutants on the land, and  to assess  effects which
are due primarily to the relatively long, sometimes circuitous  pathways
along which pollutants travel in the real world of  a  complete agricultural
watershed.
                                   85

-------
     The recommended subbasin modeling activity thus involves significantly
larger areas, or watersheds, than are currently under study in research
programs, and makes use of current and available data on land use and water
quality, supplemented to a modest extent by added water quality stations
and collection of other new data.

Development of a Master Flan for Data Bank Maintenance

     Present national priorities are very strongly directed to several is-
sues or questions which hinge on land and water qualities, and land and
water resource values. Key issues are energy production/conservation, food
production, ultimate waste disposal, hazardous and toxic substance control,
mineral resource development, disease and pest control through chemical
and biological agents, and preservation of the water resource for various
beneficial uses.

     No one will dispute the need for a system which provides for accumula-
tion, storage, and retrieval of information and data related to these issues.
This quite obviously is no simple undertaking, even with computers at hand.
Data on land characteristics, land use, and water quality are foundation
information for such a system. Accessory information of potential value
includes land used for waste disposal, pesticide use, fertilizer use, and
point source generation and dispersal of hazardous wastes, toxic substances,
and water pollutants.

     The Missouri River Basin Data Bank contains much of the above-named
information, some comprehensively, and some less than adequately. Further-
more, programs and mechanisms have been developed which permit retrieval
and versatile manipulation of stored data.

     Continuing development and use of the Data Bank should be conducted
with a two-fold objective:

     a.  To refine and extend its relevance and utility in nonpoint pollu-
tion assessment and modeling; and

     b.  In recognition of other needs and issues, to lay the groundwork
for expanding the base of information and broadening its utility.
     It is recommended, therefore, that a use-oriented study be conducted
with the objective of developing a plan for expanding the Missouri River
Data Bank and developing it to the point that it is a mature system which
meets several needs. The study should include the following:
                                   86

-------
   .  *  Assessment of the extent and availability of information,  and  whether
        provisions have been made for updating  transient  data.

     *  Development of a data needs list,  and justification  of need  through
        specification of use.

     *  Development of a list of candidate information  uses,  and  specifica-
        tion of data manipulation requirements  for these  uses.

     *  Development of data acquisition needs,  such as  added water quality
        stations or addition of parameters at existing  stations.

     *  System analysis of the above in the framework of  ADP capabilities
        and limitations, and formulation of one or more feasible  Data  Bank
        system together with plans for implementation.

     This program obviously could be structured to be very ambitious and
quite broad in scope. A sensible restriction on ambitions and scope  is
embodied in the following statement:  the Missouri River Basin Data  Bank
contains what seem to be the basic building blocks of a broadly useful
system, in a usable format; it is appropriate to obtain through further
study and analysis a definition of potential scope and  breadth of use,
and to then formulate and present development/implementation options of
varying ambition and urgency, so that we can sensibly select one  or  more
to act upon.
                                    87

-------
 APPENDIX A
DATA SOURCES
   89

-------
     The data base developed during Phase I study contained the following
categories of data which was collected from various data sources and orga-
nized in a computer retrievable/processible form.

WATER QUALITY DATA

     Water quality data were arranged according to physical, chemical,
and biological parameters. Data from the Office of Water Data Coordina-
tion (OWDG) contained information concerning number, frequency, and
location of important water quality monitoring stations.

     A large fraction of the water quality data available for the Missouri
River Basin was collected, particularly (a) those data available from
EPA's STORET system, and (b) the tabulated data from USGS between 1965
and 1970. Information from those two data sources and from other sources
                                 *i?
have been summarized by the MBIAC, and that summary has also been obtained.
Much of the water quality data is in computer-processible form, e.g., the
STORET system and the USGS system.

     Two other sources of water quality data which may contain a signifi-
cant amount of information are operating statements from both sewage treat-
ment and water treatment plants. While some portion of this information
is reported to EPA and put into STORET, most of it can be obtained from
state files or the treatment plant themselves. We did not expand our cur-
rent data bank to include these point sources.

HYDROLOGICAL DATA

     Hydrological data can be divided into the following categories:
stage, discharge, peak stage, low flow, cross section, flow duration,
flood frequency, coefficient of roughness, time of travel, and surface
inflow-outflow. The two principal sources of data on these parameters
are the USGS and the Corps of Engineers.

     The USGS publishes hydrologic data each year for each state in the
basin. We obtained their data for the 1965 to 1970 period for most states
in the basin. Besides being published in tabular form, these data are
available in computer-processible form. The USGS operates more than 1,000
surface water stations in the basin with at least 40 stations in every
basin state with the exception of Minnesota. Of these 1,000 stations,
over 900 report daily discharge, and other data less frequently. The
frequency of hydrologic data acquisition was almost always greater than
for water quality measurements.
*  Missouri Basin Interagency Committee.

                                   90

-------
     The Corps of Engineers takes  hydrologic  data  for use  in  its  reservoir
and channel straightening work. Most  of  the information obtained  is put
into the USGS computer system. The Corps also publishes flood plain studies
for specific watersheds.

     A less important source of hydrologic information for the Missouri
River Basin is the River Forecast  Center of National Oceanic  and  Atmo-
spheric Administration (NOAA) in Kansas  City. It has developed long-term
hydrographs for 800 subbasins which encompass the  entire Missouri Basin
plus an additional quarter-million square miles. These subbasins  average
about 1,000 miles2, but may be as  small  as 200  to  400 miles2. With these
hydrographs and computer models  for runoff, NOAA can successfully predict
river stages along the Missouri  River.

CLTMATOLOGICAL DATA

     Climatological factors of most importance  to  our  study are precipi-
tation and temperature. Because  of interest over a long period of time,
much data have been collected. The principal  source of these  data is  the
U.S. Department of Commerce, National Oceanic and  Atmospheric Administra-
tion, Environmental Data Service,  Asheville,  North Carolina.

     The quantity and quality of climatological data obtainable from  NOAA
are excellent and are believed to  be  sufficient for our study. A  secondary
source of data is MBIAC reports  which present climatological  information
for the Missouri River Basin in  either graphical or tabular form. These
reports were prepared from the NOAA data.

LAND USE DATA

     Land use in the Missouri River Basin can be divided  into the follow-
ing:  agriculture, recreation, urban, military, grassland, wildlife areas,
forest areas, and water areas. The sources of land use data include the
various branches of the U.S. Department  of Agriculture--the Statistical
Research Service, the Agricultural Stabilization and Conservation Service,
the Forest Service, and the Soil Conservation Service. Each state also
has a Department of Agriculture  which collects  land use data. The Depart-
ment of the Interior has land use  data in several  branches such as the
Bureau of Reclamation.
            i
     There are at least five major types of compilations  of  land  use
data. One of the most important  is the Conservation Needs Inventory.  Two
Conservation Need Inventories have been published, one for 1958  and the
other for 1967. The 1958 inventory contains data  on land use  and  conserva-
tion treatment needs by county for the entire United States.  Statistics
                                  91

-------
include cropping patterns with yield and projected yield by resource area,
soil resource group, and subbasin. The inventory information is correlated
with the soil classification data in the Missouri Basin; the correlations
are on tape and are available at the Missouri Basin Interagency Committee
Office in Lincoln, Nebraska. The inventory includes all land use except
built-up areas and land owned by the federal government.

     The 1967 inventory includes all acreage except urban and built-up
acres, and all land owned by the federal government other than cropland
operated under lease or permit. Estimates of needs for conservation treat-
ment for each major land use were based on the acreages and conditions
of the land or the vegetative cover as of the crop year 1967.

     The 1967 Conservation Needs Inventory is available on magnetic tape.
On a county basis, the data include land capability class and subclass,
information on acreage of corn and sorghum, other row crops, close grown
crops, summer fallow, total field crops, rotation hay and pasture, hay
land, conservation use only, temporarily idle cropland, total tillage ro-
tation, orchard, vineyards and bush fruit, open lands formerly cropped,
and total cropland.

     The next major source of land use information is the Census of Agri-
culture for 1969. This census includes detailed land use data by county
for each state of the union. The 1969 Census is available on tape and in-
cludes 1964 statistics for comparison with 1969 figures. For each county,
data include information about all farms, such as the number, the acreage
in the farms, and the land use. It includes the size of the farms, the
farm operator tenure, farm income and sales, farm production expenses,
machinery and equipment, amounts of livestock and poultry, and crops har-
vested. It also contains some data on irrigation and artificial drainage
as well as agricultural chemicals and commercial fertilizer use. The
crops are divided into corn, sorghum, hay, feed, field seeds, strawberries,
small grains, soybeans, peanuts, potatoes, tobacco, cotton, vegetables,
tree fruit and grapes, nursery and greenhouse products, and forest prod-
ucts.

     Another source of information on crops and land use is the Agricul-
tural Stabilization and Conservation Service of the USDA, which has a
summary by county of all farms which participate in the ASCS program.
This summary represents over 75% of all farms. Through the year 1970,
the data covered wheat, cotton, and feed grains, but the 1971 report is
more extensive—going into more detail and including other grains. ASCS
uses magnetic tapes for storing their data.
                                   92

-------
     Another data source is the detailed data  sheets  compiled by
counties for the Land Resource Appendix of  the Missouri  Basin Compre-
hensive Framework Study. These working papers  include land use  data  for
public lands which are not available in the Conservation Needs  Inven-
tory. The papers are of value because they  present  a  single  source of
data on use of these public lands.  These data  are  located otherwise  in
several government agencies having  legal jurisdiction over these  lands,
i.e., Bureau of Land Management, Forest Service, Bureau  of Reclamation,
and Corps of Engineers.

     A final source of land use information is state  cropland and live-
stock data. All states in the Missouri Basin provide  an  annual  or biannual
agricultural census and keep records of major  crop  acreages  and of the
number of livestock on farms on a county basis.  For most Missouri Basin
states, these reports have been compiled for many  years  (10  to  40 years).
While these data are not as detailed as the Conservation Needs  Inventory
or Census of Agriculture, they represent an annual data  source.

TOPOGRAPHIC DATA

     The topographic information needed for modeling  work  includes  items
such as tributary location, basin elevation, drainage area,  river mileage,
river width, river body slope, drainage area slope, etc. The primary data
source here is the USGS topographical maps  having  scales of  1:24,000 and
1:250,000, which are available for the entire  basin.  These two  types of
maps present equal elevation contour intervals of  10  ft  and  50  ft,  respec-
tively, and can be used to estimate the values of  the previously  mentioned
parameters. However, it would require a considerable  amount  of  time  and
effort to transfer this information from maps  to tables. Fortunately, ex-
cept for the drainage area slopes,  most of  the needed topographic data
have been compiled by the following agencies.

MBIAG

     A report entitled "Condensed Tabulation of River Mileage and Drainage
Areas," published in 1965, lists for approximately 900 gauged tributaries
(1) gauge location, (2) mileage above the mouth, (3)  noncontributing area,
and (4) drainage area. This publication is  a supplement  to the  1949  edi-
tion of the Corps of Engineers "Missouri River Basin-River Mileages  and
Drainage Areas."

USGS

     Gauge locations and drainage areas are always listed  in their an-
nually published "Water Resources Data." The river body slope,  stream
length, river width, drainage area, and other  characteristics of gauged
and ungauged sites are available at the USGS district offices.

                                  93

-------
SOIL CLASSIFICATION DATA

     The types of soil directly influence the quality and quantity of pol-
lutant in the surface runoff and the agricultural practices  on  the lands.
Furthermore, the agricultural practices greatly affect the water  quality
of the streams. Soil classification data, therefore,  are extremely impor-
tant in modeling work.

     A very important method of classification for the purposes of our
study was found to be based on the sediment yield of  the soil.  Based on
this parameter, the SCS delineated and classified the soil areas  into
seven categories. These data have been used to develop a soil map for all
the plains states. The map does not include those parts of the  basin in
Missouri, Iowa, and Minnesota. However, they are presently capable of com-
pleting the basin area not shown on the Great Plains  map.

     The Soil Conservation Service also has a wealth  of other data relat-
ing to soil type, land use and topography. These data are available in
soil surveys on a county-by-county basis. However, soil surveys are not
available for each county in the basin and some of them are  quite old,
but they are still unusually good information sources.

LIVESTOCK DATA

     Data on livestock are available from the State Agriculture Boards
on a county basis. Livestock inventories are taken at least  semi annually
(on 1 January and at mid-year), and livestock marketing information  is
reported monthly.

     The categories of livestock reported are as follows:

     *  Milk cows that have calved

     *  Other cattle

     *  Hogs and pigs

     *  Sheep and lambs

     *  Chickens

     *  Turkeys
                                   94

-------
     The number of cattle feedlots and the number of fed cattle marketed
are also reported, although the size of operation constituting a com-
mercial feedlot varies among states.

FERTILIZER DATA

     Information on fertilizer consumption is avail-able in great detail on
a county basis throughout the Missouri River Basin.  The tonnage of  ferti-
lizer sold (in 14 categories) is reported for about  40  different grades
and types of fertilizer. The totals are published twice yearly for  each
county.

     Fertilizer data are also collected and published annually in The
Fertilizer Supply by the U.S. Department of Agriculture's ASCS, and by
its Statistical Report Service in summary form by states. The Census of
Agriculture reports the number of farms on which fertilizers are applied
for each county, the number of acres fertilized, the tons of fertilizer
applied, and the farmers' expenditures on fertilizers.

PESTICIDES

     There are no reliable data available on a county basis regarding
pesticide applications, other than those published in the Census of Agri-
culture. The Census reports farm expenditures on a county basis for pesti-
cides for various uses, and the number of acres on which the pesticides
were used. These data, along with information obtained or obtainable from
other sources, provide a reasonably sound basis for estimating the  extent
and type of pesticide application on a county level.

Estimating Pesticide Applications

     Pesticide applications are estimated on a county level, utilizing
several different data sources and techniques:

     a.  Total production and consumption of pesticides in the U.S., as
reported annually by ASCS.

     b.  Pesticide expenditures and acres treated for each county,  as  re-
ported in the Census of Agriculture for selected years.

     c.  A mail survey of county extension agents to determine local pesti-
cide application practices.

     Much of the pesticide data was obtained by a survey of county exten-
sion agents.
                                  95

-------
    APPENDIX B
DATA ORGANIZATION
       96

-------
     The significant achievement of Phase I study was  the  compilation  of
all pertinent data in a logical fashion,  which allows easy  access  for what-
ever analysis might be desired.

     The collected data were originally  generalized on the basis  of three
separate systems:  data on water quality and hydrology come from  the stream
stations at which they were monitored; data for slope, soil classification,
land use, livestock distribution, fertilizer use, and  pesticides  use are
available by counties; and precipitation data are from each weather sta-
tion. We related all data to a common base--the watershed.  To  achieve  this,
the county-watershed relationships and county-weather  station  relation-
ships were determined and utilized to transform data  from  counties or
weather stations into the watershed. The data system which resulted di-
rectly relates water quality parameters  to physical,  economic  and clima-
tological parameters. The present file has water quality data  and hydro-
logical data for 1968, 1969, and 1970, and land use data for 1969. However,
provisions were made for updating when new data become'available.

     Two major computer files were produced for this  program.  The County
Master File (Table B-l) has one record for each of the 438 counties  in
the Missouri River Basin. Records for each county consist  of data on their
land use, fertilizer and pesticide application, soil  classification,  live-
stock distribution, topography,  and climatology. The  Station Master File
(Table B-2) contains information unique  to an individual station  and the
watershed which impacts that station. In this file, the station location,
the water quality and hydrological data, the area and  physical characteris-
tics of the watershed, and all information available  for counties in the
watershed are recorded in the County Master File are  compiled. Two hundred
thirty-three water quality stations were selected. These stations were
chosen on the basis of their locations,  and the type  and frequency of  their
measurements. Identifications and locations of these  stations  are presented
in the Phase I Final Report.!/
JL/  Vandegrift, A. E., and S. Y. Chiu, "Systems Programs for the Analysis
      of Nonurban, Nonpoint Source Pollutants in the Missouri Basin
      Region," Midwest Research Institute, Kansas City, Missouri (May
      1973).
                                  97

-------
            Table B-l.   COtJNTY FILE  INFORMATION SHEET, 1969
      State
             County_
      Watershed
      Land Resource Area
                  Land Resource Region_
 Land Characteristics:
      Soil Capability  Classes  (in acres)
   I - 	
 HE - 	
 IIW - 	
 IIS - 	
 i:..; -	
IIIE - 	
IIIW - 	
HIS - 	_
Slope Percent Class
Soil Name
IIIC -
 IVE -
 IW -
 IVS -
 IVC -
  VE -
  VW -
     VS -
     VC -
    VIE -
    VIW -
    VIS -
    VIC -
   VIIE -
    VIIW -
    VIIS -
    VHC -
    VIIIE -
    VIIIW -
    VIIIS -
    VIIIC -
                          Soil Erosion Class
Total Precipitation
                      (Year)
       in.
           (Spring)
Rainfall Intensity
  in. 	in.
      (Summer)
(Fall)
                                                in/30 min
        in.
         in.
(Winter)
 Land Use (in Sq.  Miles)
       Total Land Area_
       Land in Farm
       Other Land Area
           Harvested Crop_
           Pasture Crop	
           Other Cropland_
           Total Cropland_
           Woodland	
           Irrigated	
                                  Other Farmland
  Crop Data (in Sq. Miles)
      Corn	
      Sorghums
      Wheat
      Other Small Grain
      Soybeans 	
      Hay	
      Cotton 	
      Peanuts	
                         Tobacco 	
                         Potatoes 	
                         Vegetables 	
                         Berries 	
                         Orchards
                         Other Crops 	
                         Greenhouse (Sq. Ft.)
                                   98

-------
                      Table B-l.   (Concluded)
Fertilizer (tons):
                  1st Half Year
2nd Half Year
     Nitrogen
     P2°5
     K20
                              Annual
     Lime
     Herbicide Applied
     Insecticide Applied
     Fungicide Applied

Livestock and Poultry;
Cattle and Calves  (total)
  Cattle Feedlots  (600+)
  Cattle in Feedlots  (600+)

Hogs  (total)  	
          Sheep (Total)  	
          Horses and Ponies
          Chickens (Total 3 Mo+).
   Hoglots  (200+) 	
   Hogs in Lots  (200+)
             Chicken houses (1600+)	
             Chickens in houses (1600+)
                                    99

-------
                    Table B-2.   STATION FILE INFORMATION SHEET,  1969
     Location;
          Stream Name
               Location
          MRI Station Number
          State
                     OWDC Sub-Sub-Basin
County_
Latitude
Longitude_
          STORET Station__
          OWDC SW Station
         STORET Agency_
        USGS Station
               OWDC WQ Station.
          Total Watershed Area
                  Downstream Station
          Distance to Downstream Station
     Land Characteristics:
   Main Channel  Slope
Soil Erosion Class

Soil Capability Classes (in
I - IIIC
HE -
IIW -
IIS -
IIC -
HIE -
IIIW -
HIS -
Rainfall Intensity
Total Precipitation

IVE
IVW
IVS
IVC
VE
VW

(Year)
Soil Infiltr
acres)
-
-
-
-
••
-
in/30 min
in.
(Spring)
   ft/mi  Slope Percent Class
                                                  VS -
                                                  VC -
                                                 VIE -
                                                 vna -
                                                 VIS -
                                                 VIC -
                                                VIIE -
                                                              in/hr  Soil Index
                                           VIIW
                                           VIIS
                                           VIIC
                                          VIIIE
                                          VIIIW
                                          VIIIS
                                          VIIIC
                                                           in
                                              in.
                                                  (Summer)
                                £n.
                                     (Fall)
                            in.
                                (Winter)
                       in.
:ation Data;   (Averages of Year,  Spring,  Summer, Fall,  and Winter)
   Temp.
   D.O.
   BOD
   Total  N
   T.D.S.
   N03
   P04
   T.S.S.
                S04
                CL
                T. Coliform
                F. Coliform
                Herbicides
                Insecticides
                Fungicides
                Flow
                                        100

-------
                        Table B-2.  (Continued)
Land Use:  (Sq. Miles)

      Total Land Area _
      Land in Farm
      Other Land Area
Crop Data;  (Sq. Miles)
     Corn
     Sorghums_
     Wheat
     Other Small Grain.
     S oybeans	
     Hay	
     Cotton
     Peanuts
Fertilizer Data;  (Tons)
   Harvested Crop
   Pasture Crop	
   Other Cropland_
   Total Cropland^
   Woodland
   Irrigated
   Other Farmland
          Tobacco	
          Potatoes	
          Vegetables_
          Berries	
          Orchards
           Other Crops
           Greenhouses  (Sq. Ft.)
               1st Half Year
2nd Half Year
     Nitrogen
     P205
     K20
                             Annual
     Lime
     Herbicide Applied
     Insecticide Applied
     Fungicide Applied
                                  101

-------
                           Table B-2.  (Concluded)
Livestock and Poultry;

     Cattle and Calves  (Total).
       Cattle Feedlots  (600+)
Sheep (Total)	
Horses and Ponies
       Cattle in Feedlots  (600+)
     Hogs  (Total),
Chickens (Total 3 Mo.+).
       Hoglots  (200+)	
       Hogs in  Lots  (200+).
  Chicken Houses (1,600+)	
  Chickens in Houses (l,600+)_
Contributing Counties;(Up to 100 Counties)
     State/County Code.
     Total County Area_
County Code_
     County Area to This Station_
     Distance from County to Station_
                                    102

-------
        APPENDIX C
DATA PROCESSING TECHNIQUES
            103

-------
     We have developed the data processing procedures by converting the
computer program originally developed for use on a Burroughs computer to
be used on CDC/6600 System.

     The conversion was needed because the available statistical package
could not be used on the Burroughs computer, which is relatively slow and
thus expensive. Direct conversion is not feasible due to dissimilarities  of
the two systems. Consequently, the data bank was transformed from Burroughs
tape to an intermediate tape on IBM/360 from which the data are transformed
onto a CDC/6600 System tape.

     As it is expensive to run the CDC/6600 tape each time the data are
processed, a separate data file was created which contains only selected
parameter inputs on CDC/6600 requiring shorter core storage space. Backup
data files, duplicates of original tapes, were created to avoid accidental
loss of data.

     The reduced version Data Bank contains essential information which
can be automatically tabulated in data sets for each of the water quality
monitoring stations located on discrete segments of the river tributary
system. In this way, we can correlate the water quality vs land use param-
eters and other environmental factors by using standard statistical pack-
ages.

     The data processing sequence, from the existing data bank on Burroughs
3500 tape to statistical analysis of data, is indicated in Figure C-l. The
following sequential steps related to the creation of new data files were
completed:

     1.  Conversion of the station master and county master tapes from
B3500 to IBM/360, using the existing program.

     2.  Writing a 360 COBOL Program to convert the IBM/360 station master
tape to an unpacked format acceptable to the CDC/6600.

     3.  Creation of the unpacked station master tape.

     4.  Writing a 360 COBOL Program to read the full station master tape
and produce a reduced station data tape.

     5.  Creation of the reduced station data tape.

     6.  Development of programs to select data from the reduced tape for
appropriate statistical analyses.

     From unpacked data files created from the original computer tapes,
data were reassembled ±n tabular form.

                                   104

-------
Tape Conversion Sequence
               CONV #1
                                CONV #2
Data Reduction Sequence
  BCD
   L
REDUCE
#1

y
REDUCE
#2

\
'WORKING [
1 FILE \
\ \
                             PARAM
                             SELECTION
Data Analysis Sequence
(1) SPSS

                    WORKING
                    FILE
                              SPSS
                                                    REPORTS
                      SPSS
                      CARDS
(2) Fortran
                   WORKING
                   FILE
                    PROGRAM
                            PROGRAM
                     Figure C-l.  Data processing sequence
                                     105

-------
    APPENDIX D
COMPUTER PROGRAMS
        106

-------
     Automatic data processing for the current  Phase II(A)  study involved
development of a number of computer programs. The system language used
for these programs was COBOL,  to be consistent  with computer  programs  de-
veloped during Phase I. These  programs are adaptable for FORTRAN language
subroutines, and other scientific subroutine packages available  commer-
cially.

     The following programs were developed during this study:

     1.  Conversion of Burroughs 3500 Tape of  Station Master  File to IBM/
360 tape.

     2.  Creation of UNPACK file to unpack IBM/360 Station Master File.

     3.  Conversion of B3500 County Master File to IBM/360 tape.

     4.  Creation of a reduced Station Master File (REDUCE, CNVRT1,  and
CNVRT2).

     5.  "SORT" file to read 13 subbasin files.

     6.  "CNVRT3" file to add annual rainfall  data to the file.

     7.  Sample program of SPSS Library subroutine for regressions and
correlations.

     8.  Use of subfiles of SPSS to get tables  showing coefficient of
variation between predicted and observed values of dependent  variables
in multiple regression equations for each subbasin.

     9.  Print tables of independent variables  and dependent  variables
giving correlation and regression analysis.

    10.  Program "SBASIN" to list tables of stations with valid  observa-
tions for variables being analyzed.

    11.  Program to create a file (BDS 3807) from total Station  Master
with data to be analyzed and store tables on disk.

    12.  Program to print tables of all station values for variables to
be analyzed.
                                  107

-------
1.   CONVERSION OF BURROUGHS TAPE TO IBM
            (STATION MASTER)
                  108

-------
// JOB
// OPT
FIRST
ION LOG,L
CONVERT H^bOO STATION M
INK
// EXEC ASSEMBLY

V218
RO
Rl
K2
R3
R4
R5
R6
R7
RR
R9
KlO
Rll
R12
R13
R14
R15

















GETIN


»
*











TITLE
START
EUU
EWU
EUU
EQU
FUU
EQU
EQU
EQU
FQU
fQU
EQU
EUU
EQU
F.OIJ
EQU
EQU
USING
RALR
BCTR
t3CTR
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
OPEN
GET
LA
LA
BAL
LA
LA
BAL
LA
LA
BAL
LA
LA
6AL
LA
LA
BAL

-------
TAPEOF
LA
LA
•SAL
LA
LA
BAL
LA
LA
BAL
LA
LA
BAL
LA
LA
BAL
LA
LA
BAL
PUT
R
EQU
CLOSE
EOJ
R9
Rl
HI
R9
Rl
Rl
R9
HI
Rl
R9
Rl
Rl
R9
Rl
Rl
R9
Rl
Rl
•PL6403
0,640
1,MVOLOOP6
»PL?>13
0,52
1,MVOLOOP*
,PL6404
0,640
1,MVOLOOP6
,PL514
0,52
1 ,MVOLOOP4
,PLb40b
0,640
l,MVOLOOPb
•PL515
0,451
1,MVOLOOP4
STOUT
GFTIN
*
ST


IN, STOUT

MVOLOOP4
EQU
MVO
01
LA
BCT
BR
               0(4,H9),0(4,R9)
               3(R9),X
-------
MVOLOOP6
NEG
NEXT
EOU
MVC
MVO
NI
TM
BO
01
8
01
LA
SAVE,0
-------
2.  UNPACK IBM STATION MASTER
             112

-------
// JOB UNPACK   UNPACK THE IBM-360 STATION MASTER FILE
ALLOC F1=OK,F2=OK
// OPTION LOG»LINK
// ASSGN SYS004»X<130<
// EXEC FCOBOL
       IDENTIFICATION DIVISION.

       PROGRAM-ID. 
-------
IT  I                               PICTURE 9(3) COMP.
77  J                               PICTURE 9(3) COMP.

01  STATION-RhC-OUT.
    03  STATION-IDENT-OUT           PICTURE X(15M.
    03  STATION-DATA-OUT.
        OS  SD-OUT                  PICTURE 9(8) OCCURS 40.
    03  STATION-DATA-HY-YfcA»-OUT    OCCURS b.
        OS  SDBY-1-OuT              PICTURE 9(6).
        05  SOBY-?-OUT.
            07  SDBY-2-OUT-C        PICTURE 9(12) OCCURS 640.
        05  SOBY-2-OUT-R            REDEFINES SDMY-2-OUT.
            07  SDBY-P-OUT-D        OCCURS 640.
            09  SDBY-2-OUT-SIGN     PICTURE X.
            09  FILLER              PICTURE X(ll).
        OS  SDBY-3-OUT              PICTURE 9(8) OCCURS 51.
    01  COUNTIES-IN-STATION-OUT.
        05  CIS-OUT                 PICTURE 9(8) OCCURS 400.
    03  FILLER                      PICTURE X(4).
01  REC-OUT  REDEFINES STATION-REC-OUT.
    03  PRU-OUT                     PICTURE X(J6«0) OCCURS 12.

PROCEDURE DIVISION.

AOOO-CONTROL SECTION.

A001.
    NOTE    THIS SECTION CONTROLS THE EXECUTION
            OF THE PROGRAM.

AOIO-START.
    OPEN INPUT  STATION-FILE-INPUT.
    OPEN OUTPUT STATION-FILE-OUTPUT.

A020-READ-INPUT.
    READ STATION-FILE-INPUT
        AT END GO TO A030-STOP.
    PERFORM COOO-UNPACK-AND-WRITE.
    GO TO A020-READ-INPUT.

A030-STOP.
    CLOSE STATION-FILE-INPUT.
    CLOSE STATION-FILE-OUTPUT.
    STOP RUN.

COOO-UNPACK-AND-wRITE SECTION.

COOl.
    NOTE    THIS SECTION BUILDS AND WRITES ONE
            STATION-FILE-OUTPUT LOGICAL RECORD.

C010-MOVE-ALPHA.
        MOVE STATION-IUENT-IN TO STATION-IDtNT-OUT.

C020-MOVE-COMP.
    PERFORM C030-MOVE-STATION-DATA
            VARYING  I FROM 1 BY 1 UNTIL I GREATER THAN 40.
    PERFORM C040-MOVE-BY-YEAR
            VARYING  I FROM 1 BY 1 UNTIL I GREATER THAN 5.
    PERFORM C070-MOVE-COUNTIES
            VARYING  I FROM 1 BY 1 UNTIL I GREATER THAN 400.
    GO TO ClOO-rfRITE-OUTPUT.

C030-MOVE-STATION-DATA.
    MOVE SD-IN (I) TO SD-OUT (I).

                              114

-------
       C040-MO\/E-HY-YEAK.
           MOVE SDHY-1-IM  (I)  TO  SUnY-1-OUT  (I).
           PERFORM COSO-r-'OvE-WATtK-OATA
                   VARYING J  FROM  1  BY  1  UNTIL  J  GREATER  THAN 640,
           PERFORM cobo-MOVE-LANo-DATA
                   VARYING J  FROM  1  BY  1  UNTIL  J  GREATER  THAN 51.

       C050-MOVE-WATER-DATA.
           MOVE SDBY-2-IN  (I  J)  TO SDHY-2-OUT-C (I  J).
           IF SDBY-2-IN  (I J)  LESS THAN  ZERO
               MOVE <-<  TO SD8Y-2-OUT-SIGN (I  J).

       C060-MOVE-LAND-DATA.
           MOVE SDBY-3-IN  (I  J)  TO SD6Y-3-OUT  (I  J).

       C070-MOVE-COUNTIES.
           MOVE CIS-IN  (I) TO CIS-OUT  (I).

       C090-WRITE-PRU.
           MOVE PPU-UUT  (I)  TO OUT-RECORD.
           WHITE STATIOfJ-RECORD-OUT.

       C100-WRITE-OUTPUT.
           PERFORM C090-WRITE-PRU
                   VARYING I  FROM  1  BY  1  UNTIL  I  GREATER  THAN 12.

/*
// EXEC LNKEDT
// ASSGN SYS001«X
-------
3.  CONVERT COUNTY MASTER TO IBM
              116

-------
// JOM THlK
-------
4,  CREATE A REDUCED STATION MASTER FILE
                  118

-------
CONVRT,T300,CM 10000.
ACCOUNT,M350415,MIRWEST,3807C.
REQUEST»TAPEI»HY,X,C=36HO.  /b40b/ NO RING
SKIPF,TAPEI,1.
RFL,50000.
COBOL.
LGO.
SAVE,RTAPE.
RETURN,TAPEI.
REWIND,RTAPE.
REWIND,LGO.
RFL,50000.
COBOL.
LGO.
SAVE,TAPEX.
REWIND,TAPEX.
REWIND,LGO.
RFL,50000.
FTN.
LGO.
SAVE,STNFIL.
REQUEST,TAPED,HY.         /2252X RING IN
REWIND,STNFIL,TAPEO.
COPYBF,STNFIL,TAPED,!.
RETURN,TAPED.
UNSAVE,RTAPE.
UNSAVE,TAPEX.
RFL,10000.
GET,COST(LI8RARY)
COST.
EXIT.
RFL,10000.
GET,COST(LIBRARY)
COST.
&
       IDENTIFICATION DIVISION.

       PROGRAM-ID. "REDUCE"

       AUTHOR.     RAY POSCH
                   MIDWEST RESEARCH INSTITUTE
                   KANSAS CITY, MO.

       REMARKS.    THIS PROGRAM READS THE UNPACKED  STATION  MASTER  FILE
                       AND CREATES A REDUCED STATION  DATA FILE.

       ENVIRONMENT DIVISION.

       CONFIGURATION SECTION.
       SOURCE-COMPUTER. 6600.
       OBJECT-COMPUTER. 6600.

       INPUT-OUTPUT SECTION.

       FILE-CONTROL.
           SELECT MASTER-STATION-FILE  ASSIGN  TO TAPEI.
           SELECT REDUCED-STATION-FILE ASSIGN  TO RTAPE.
           SELECT PRINT-FILE           ASSIGN  TO OUTPUT.

       DATA DIVISION.
                              119

-------
FILE SECTION.
FO  PRINT-FILE
    LAHEL RECURU  IS OMITTED
    RECORD CONTAINS 132 CHARACTERS
    DATA RECORD IS PRINT-LINE.

01  PRINT-LINE.
    03  CARRIAGE-CONTROL
    03  PRINT-DATA
                                    PICTURb X.
                                    PICTURE XU31)
FI>  MASTER-STATION-FILE
    LABEL RECORDS  ARE  OMITTED
    RECORD CONTAINS  3680  CHARACTERS
    DATA RECORD  IS STATION-RECORD-IN.
01  STATION-RECORU-IN.
    03  RECORD-IN


FD  REDUCED-STATION-FILE
    LABEL  RECORDS  AR£  OMITTED
    RECORD CONTAINS  2100  CHARACTERS
    DATA RECORD  IS STATION-RECORD.
                                    PICTURE X(3b«0)
                                     PICTURE  9999.
                                     PICTURE  99.
                                     PICTURE  99.


                                     PICTURE  9999.
                                     PICTURE  99.
                                     PICTURE
01  STATION-RECORD.
    03  STATION-IDENTIFICATION.
        05  STATION-NUMBER          PICTURE SMrf).
        05  STATE-NAME              PICTURE X(16).
        05  COUNTY-NAME             PICTURE X(16).
        05  STREAM-NAME             PICTURE X(16).
        05  LOCATION-NAME           PICTURE X(16).
        05  LATITUDE.
            07  DEGREES-LATITUDE
            07  MINUTES-LATITUDE
            07  SECONDS-LATITUDE
        05  LONGITUDE.
            07  DEGREES-LONGITUDE
            07  MINUTES-LONGITUDE
            07  SECONDS-LONGITUDE   PICTURE 99.
    03  STATION-DATA.
        05  TOTAL-AREA              PICTURE 9(8).
        05  MAIN-CHANNEL-SLOPE      PICTURE 9(5)V9<3).
        05  RAIN-INTENSITY          PICTURE 9(5)V9(3).
        05  RAIN-EROSION-INDEX      PICTURE 9(5)V9(3).
        05  SLOPE-PCT-CLASS         PICTURE 9<5)V9(3).
        05  EROSION-CLASS           PICTURE 9(5)V9(3).
        05  SOIL-INFILTRATION       PICTURE 9(5)V9(3).
        05  SOIL-INDEX              PICTURE 9(5)V9(3).
        05  SOIL-CLASS-DATA.
            07  SOIL-TYPE           OCCURS 8.
                09  SOIL-CLASS-AREA PICTURE 9(8) OCCURS 4.
        05  SOIL-RESOURCE-DATA.
            07  SOIL-RESOURCE-TYPE  OCCURS 8.>
                09  SOIL-RESOURCE-GROUP PICTURE 9(8) OCCURS 4,
    03  STATION-DATA-HJY-YEAR.
        05  DATA-YEAR               PICTURE 9(8).
        05  WATER-QUALITY.
            07  WATER-QUALITY-PERIOD  OCCURS 5.
                 09   TEMPERATURE-F
                 09   STREAMFLOW-CFS
                 09   TURBIDITY-JU
                 09   DISSOLVED-OXYGEN-MGL
                 09   BIO-OXY-DEMAND-MGL
                 09   TOTAL-PHOSPHORUS-MGL
                 OQ   TOTAI -HOI TFORMl-innMI
                                            PICTURE S9(9)V9(3)
                                            PICTURE S9(9)V9(3)
                                            PICTURE S9(9)V9(3)
                                            PICTURE S9(9)V9(3)
                                            PICTURE S9(9)V9(3)
                                            PICTURE S9(9)V9(3)
                                            PTPTIIBF
                             120

-------
    05  STREAM-NAME-IN
    05  LOCATION-NAME-IN
    05  FILLER
03  STATION-DATA-IN.
    05  TOTAL-AREA-IN
    Ob  MAIM-CHANNEL-SLOPE-IN
    05  RAIN-INTfNSITY-IN
    05  SLOPt-PCT-CLASS-lN
    05  EROSION-CLASS-IN
    05  SOIL-INFILTPATION-IN
    05  SOIL-INUEX-IN
    05  FILLKR
    05  SOIL-CLASS-DATA-IN
03  STATION-uATA-HY-YEAR-IN
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
OCCURS
X
X
X
9
9
9
9
9
9
9
X
X
b.
(16) .
(16) .
(56) .
(H)
)
(5)
(5)
(b)
(b)
(b)
(6)
•
V9
V9
V9
V9
V9
V9
•

(3)
(3)
(3)
(3)
(3)
(3)

(256) .



                                PICTURE
05  DATA-YEAR-IN
05  WATFK-WJALITY-IN.
    07  i«ATE*-UUALlTY-PERIODS-IN  OCCURS 5.
        U9  STAT-TYPE-IN    OCCURS 64.
            11  STATISTIC-1 PICTURE S9(9)V9(3>.
            11  STAT-1 REDEFINES STATISTIC-1.
                13  STAT-SGN-1  PICTURE  X.
                13  STAT-VAL-1  PICTURE  9(b)V9(3)
            11  STATISTIC-2 PICTURE S9(9)V9(3).
05  LAND-IISE-CROPS-IN.
    05
    05
    05
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
LAND-ARE.A-IN
LAND-IN-FARMS-IN
OTHER-LAND-IN
TOTAL-CROP-AREA-IN
PICTURE
PICTURE
PICTURE
PICTURf
4(8) .
9(8) .
9(8) .
9(8) .
CHOPLAND-HARVESTED-IN PICTURE 9(8)
PASTURE-GRAZING- IN
OTHER-CROPLAND- IN
wOODLANCJS-IN
IRRIGATEO-IN
OTHER-FARMLAND- IN
CORN
SORGHUM
*HEAT
OTHER-GRAIN
SOYBEANS
HAY-IN
COTTON
PFANUTS
TOBACCO
POTATOES
VEGETABLES
«ERRIES
ORCHARDS
OTHER-CROPS
GREENHOUSES
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
9(H) .
9(8) .
9(8) .
9(8) .
9(8) .
9(8).
9(8) .
9(8).
9(8) .
9(8) .
9(8) .
9(8) .
9(8) .
9(8).
9(8) .
9(8) .
9(8) .
9(8) .
9(8) .
9(8) .
LIVESTOCK-IN.
07
07
07
07
07
07
07
07
CATTLE-1N
FILLER
HOGS-IN
FILLER
SHEEP-IN
FILLER
CHICKENS-IN
FILLtR
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
PICTURE
9(8) .
X(16) .
9(8) .
X(16) .
9(8) .
X(8) .
9(8) .
X(16) .
FERTILIZER-IN.
07



FERTILIZER-PERIODS
09 NITROGEN-IN
09 PHOSPHATE-IN
09 POTASH-IN
OCCURS
PICTURE
PICTURE
PICTURE
2.
9(8) .
9(8).
9(8) .
PESTICIDE-IN.
07
07
FILLER
HFRRTrrnp-TN
PICTURE
PTTTIIRF
X(8) .
9f ft) .
                       121

-------
                09  TOTAL-COLIFORM2-
                09  FILTERED-RSD-M6L
                09  DISSOLVED-SOLIDS
                09  NITRATE-MGL
                09  HERBICIDE-TY
                09  INSECTICIDE-TY
                09  FUNGICIDE-TY
                09  UNUSED-1
                09  UNUSEO-2
                09  UNUSEU-3
                09  UNUSED-4
        05  WATER-STATISTICS
            07  STATISTICS-PERIOD
                09  WATER-STATISTIC
        05  LAND-USE.
            07  LAND-AREA
            07   LAND-IN-FARMS
            07   OTHER-LAND
        05  FARM-LAND.
            07  TOTAL-CROP-AREA
            07   CROPLAND-HARVESTED
            07   PASTURE-GRAZING
            07   OTHER-CROPLAND
            07  IRRIGATED
            07  OTHER-FARMLAND
        05  CROPS.
            07  ROW-CROPS
            07  SMALL-GRAIN
            07  HAY
            07.  WOODLANDS
            07  C-FACTOR
        05  LIVESTOCK.
            07  CATTLE
            07  HOGS
            07  SHEEP
            07  CHICKENS
        05  FERTILIZER.
            07  NITROGEN
            07  PHOSPHATE
            07  POTASH
        05  PESTICIDE.
            07  HERBICIDE
            07  INSECTICIDE
            07  FUNGICIDE
        05  PRECIPITATION.
            07  RAINFALL
    03  FILLER
01  SEGMENTED-RECORD
    03  RECORD-DATA

WORKING^STORAGE SECTION.
77
77
77
77

01
I
J
K
S-VALUE
USAGE COMP
USAGE COMP
USAGE COMP
STATION-REC-IN.
03  STATION-IDENT-IN.
    05  STATION-NUMBER-IN
    05  FILLER
    05  STATE-NAME-IN
    05  COUNTY-NAME-IN
    05  LATITUDE-IN
    0=;  I ONRTTimF-Tl\i
                                100ML   PICTURE
                                        PICTURE
                                -TPD    PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                        PICTURE
                                REDEFINES WATER
                                OCCURS b.
                                PICTURE S9(9)V9

                                PICTURE 9(8).
                                PICTURE 4(8).
                                PICTURE 9(8).
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                S9(9)V9
                                59(9)V9
                                S9(9)V9
                                •QUALITY
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                       (3)
                               (3) OCCURS 1«.
                                PICTURE
                                PICTURE
                                PICTURE
                                PICTURE
                                PICTURE
                                PICTURE
                        9(8) .
                        9(8) .
                        9(8) .
                        9(8) .
                        9(8) .
                        9(8) .
                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(5)V9(3) OCCURS 4.

                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).

                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).

                                PICTURE 9(8).
                                PICTURE 9(8).
                                PICTURE 9(8).

                                PICTURE 9(8) OCCURS 5.
                                PICTURE X(92).

                                REDEFINES  STATION-RECORD.
                                PICTURE X(100)  OCCURS 21.
PICTURE 99.9.
PICTURE 999.
PICTURE 999.
PICTURE S9(9)V9(3)
                                     PICTURE
                                     PICTURE
                                     PICTURE
                                     PICTURE
                                     PICTURE
                                     PTPTIIBP
                        9(8) .
                        X(12)
                        X(16)
                        X(16)
                        9(8).
                        Q
                              122

-------
            07  1NSECTICIOE-1M
            07  H)NGICIDE-IN
        05  PHECIPITATION-IN
    03  COUNTIES-IN-STATION-IN
    03  FILLi-R

01  STATION-LOGICAL-REC
    03  PRU-IN

PROCEDURE DIVISION.

AOOO-CONTPOL SECTION.
                        PICTURE
                        PICTURE sra).
                        PICTURE x(40).
                        PICTURE x<32oo>.
                        PICTURE x(4>.

                        REDEFINES STATION-REC-IN.
                        PICTURE x<36«0) OCCURS i?,
AOOl.
    NOTE
THIS SECTION CONTROLS EXECUTION
OF THE PROGRAM.
A010-START.
    OPEN INPUT  MASTER-STATION-FILE WITH NO REWIND.
    OPEN OUTPUT REOUCED-STATION-FILE.
    OPEN OUTPUT PRINT-FILE.
    MOVE SPACES TO PRlNT-uATA.

A030-READ-RECOPD.
    PERFORM A050-READ-PRU
        VARYING I FROM 1 BY 1 UNTIL I GREATER THAN 12.
    PERFORM COOO-REOUCE-DATA.
    GO TO A030-READ-RECORD.

A050-READ-PRU.
    READ MASTER-STATION-FILE
        AT END GO TO A090-STOP.
    MOVE RECORD-IN TO PRU-IN  (I).

A090-STOP.
    CLOSE MASTER-STATION-FILE.
    CLOSE REDUCED-STATION-FILE.
    CLOSE PRINT-FILE.
    STOP RUN.

COOO-REDUCE-DATA SECTION.
C001.
    NOTE
THIS SECTION BUILDS THE REDUCED
STATION DATA RECORDS.
C010-REDUCTION-CONTROL.
    MOVE SPACES TO STATION-RECORD.
    PERFORM C030-MOVE-IDENT.
    PERFORM C050-MOVE-STATION-DATA.
    PERFORM C070-MOVE-BY-YEAR.
    PERFORM EOOO-OUTPUT.
    GO TO C900-EXIT.

C030-MOVE-IDENT.
    MOVE STATION-NUMBER-IN TO STATION-NUMBER.
    MOVE STATE-NAME-IN
    MOVE COUNTY-NAME-IN
    MOVE STREAM-NAME-IN
    MOVE LOCATION-NAME-IN
    MOVE LATITUOE-IN
    MOVE LONGITUOE-IN
               TO STATE-NAME.
               TO COUNTY-NAME.
               TO STREAM-NAME.
               TO LOCATION-NAME.
               TO LATITUDE.
               TO LONGITUDE.
C050-MOVE-STATION-DATA.
    MOVE TOTAL-AREA-IN         TO TOTAL-AREA.
    MOV/F MATN-C.HANNFI -SI OPF-TN TO MATN-CHANNFI -«;i OPF.
                           123

-------
    MOVE PAIN-INTFNSITY-IN
    MOVE SLOPE-PCT-CLASS-IN
    MOVE EROSION-CLASS-1N
    MOVF SOIL-INFILTRATION-IN
    MOVE SOIL-INOEX-IN
    MOVE SOIL-CLASS-DATA-IN
                           TO RAIN-INTENSITY.
                           TO SLOPE-PCT-CLASS.
                           TO EROSION-CLASS.
                           TO SOIL-INFILTRATION.
                           TO SOIL-INDEX.
                           TO SOIL-CLASb-DATA.
C070-MOVE-BY-YEAR.
    MOVE 2 TO I.
    MOVE OATA-YEAR-IN  (I) TO DATA-YEAR.
    PERFORM C090-MOVE-WATER-DATA
        VARYING J FROM  1 BY 1 UNTIL J  GREATtR  THAN 5.
    MOVE LAND-AHEA-IN  (I)
    MOVE LAND-IN-FARMS-IN  (I)
    MOVE OTHER-LANO-IN  (I)
    MOVE TOTAL-CKOP-AR£A-IN  (I)
                               TO LAND-AREA.
                               TO LANU-IN-FARMS.
                               TO OTHER-LAND.
                               TO TOTAL-CROP-AREA.
    MOVE CROPLAND-HARVESTED-IN  (I)  TO  CROPLAND-HARVESTED.
    MOVE PASTURE-GRAZING-IN  (I)
    MOVE OTHER-CROPLAND-IN  (I)
    MOVE WOODLANOS-IN  (I)
    MOVE IRRIO.ATED-IN  (i)
    MOVE OTHER-FARMLAND-IN  (I)
    ADD CORN  (I)     SORGHUM  (I)
        PEANUTS  (I)  TOBACCO  (I)
        ORCHARDS  (I)   OTHER-CROPS  (I)
            GIVING ROW-CROPS.
    ADD WHEAT  (I)    OTHER-GRAIN  (I)  GIVING  SMALL-GRAIN.
                               TO PASTURE-GRAZING.
                               TO OTHER-CROPLAND.
                               TO WOODLANDS.
                               TO IRRIGATED.
                               TO OTHER-FARMLAND.
                                SOYBEANS  (I)    COTTON (I)
                                POTATOES  (I)  VEGETABLES  (I)
    MOVE HAY-IN  (I)
    MOVE CATTLE-IN  (I)
    MOVE HOGS-IN  (I)
    MOVE SHEEP-IN  (I)
    MOVE CHICKENS-IN  (I)
                     TO HAY.
                     TO CATTLE.
                     TO HOGS.
                     TO SHEEP.
                     TO CHICKENS.
ADD NITROGEN-IN (I 1) NITROGEN-IN  (I 2) GIVING NITROGEN.
ADD PHOSPHATE-IN  (I 1) PHOSPHATE-IN  (I 2) GIVING PHOSPHATE.
ADD POTASH-IN (I  1) POTASH-IN  (I 2) GIVING POTASH.
MOVE HERBICIDE-IN  (I)   TO HERBICIDE.
MOVE INSECTICIDE-IN (I) TO INSECTICIDE.
MOVE FUNGICIDE-IN  (I)   TO FUNGICIDE.
MOVE PRECIPITATION-IN  (I) TO PRECIPITATION.
C090-MOVE-WATER-DATA.
    PERFORM  C110-CONVERT-
         VARYING  K  FROM
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  STATISTIC-1
    MOVE  ZERO  TO UNUSED-1
ERT-VALUE
M 1 BY 1
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
(I J
ED-1
ED-3
02)
03)
04)
07)
09)
32)
42)
43)
51)
52)
56)
62)
63)
64)
(J)
(J)
                          UNTIL K GREATER THAN 64.
                          TO TEMPERATURE-F  (J).
                          TO STREAMFLOW-CFS  (J).
                          TO TURBIDITY-JU (J).
                          TO DISSOLVED-OXYGEN-MGL  (J).
                          TO BIO-OXY-DEMAND-MGL  (J).
                          TO TOTAL-PHOSPHORUS-MGL  (J).
                          TO TOTAL-COLIFORM1-100ML (J),
                          TO TOTAL-COLIFORM2-100ML (J),
                          TO FILTERED-RSD-MGL  (J).
                          TO DISSOLVED-SOLIDS-TPD  (J).
                          TO NITRATE-MGL  (J).
                          TO HERBICIDE-TY (J).
                          TO INSECTICIDE-TY  (d).
                          TO FUNGICIOE-TY (J)-
                           UNUSED-2  (J)
                           UNUSED-4  (J).
C110-CONVERT-VALUE.
     IF STAT-SGN-1  (I  J  K)  EQUALS "-"
        MULTIPLY STAT-VAL-1  (I  J K)  BY -1.0 GIVING S-VALUE
     ELSE MULTIPLY  STAT-VAL-1  (I J K)  BY +1.0 GIVING S-VALUE.
     MOVE S-VALUE TO STATISTIC-1 (I J  K).
                             124

-------
C900-EXIT.
    EXIT.


FOOO-OUTPUT SECTION.
 E001.

     NOTE
            THIS SECTION CONTROLS THE WRITING OF THE RECORD
            TO THE NEW FILE AND PRINTING OF THE DATA.
 EOlO-OUTPUT-CUNTnOL.
     WRITE STATION-RECORD.
     PERFORM E030-PRINT-LINE
         VARYING I  FROM 1  BY  1  UNTIL  I  GREATER  THAN  21.

 E030-PRINT-LINE.
     MOVE RECORD-DATA  (I)  TO  PRINT-DATA.
     IF  I GREATER  THAN 1
             MOVE  SPACE TO  CARRIAGE-CONTROL
         ELSE MOVE  0 TO CARRIAGE-CONTROL.
     WRITE PRINT-LINE.

 IDENTIFICATION DIVISION.
 PROGRAM-ID. "CNVRT1"
 AUTHOR.      RAY POSCH
             MIDWEST RESEARCH INSTITUTE
             KANSAS CITY.  MO.
 REMARKS.    THIS  PROGRAM  READS THE COBOL  FORMAT REDUCED STATION
                 MASTER TAPE  AND CREATES THE INTERMEDIATE
                 FORTRAN COMPATIBLE TAPE.
 ENVIRONMENT DIVISION.
 CONFIGURATION SECTION.
 SOURCE-COMPUTER.  6600.
 OBJECT-COMPUTER.  6600.
 INPUT-OUTPUT SECTION.
 FILE-CONTROL.
     SELECT STATION-MASTER-C   ASSIGN  TO RTAPE.
     SELECT STATION-MASTER-F   ASSIGN  TO TAPEX.
     SELECT PRINT-FILE        ASSIGN  TO OUTPUT.
 DATA DIVISION.
 FILE SECTION.
 FD   PRINT-FILE
     LABEL RECORD  IS OMITTED
     RECORD CONTAINS 132 CHARACTERS
     DATA RECORD IS PRINT-LINE.
 01   PRINT-LINE.
     03   CARRIAGE-CONTROL
     03   PRINT-DATA
 FD   STATION-MASTER-C
     LABEL RECORD  IS OMITTED
     RECORD CONTAINS- 2100  CHARACTERS
     DATA RECORD IS STATION-RECORD-C
-01   STATION-RECORD-C.
                                    PICTURE X.
                                    PICTURE X(131).
03  SEG-1.
    05  DATA-l-A
    05  DATA-l-B
    05  DATA-l-C
    05  DATA-l-D
    05  DATA-l-E
    05  DATA-l-F

03  SEG-?.
    05  DATA-2-A
    05  DATA-2-B
    05  SEG-2-C.
        07  DATA-2-C

03  SEG-3.
                                     PICTURE
                                     PICTURE
                                     PICTURE
                                            9(a).
                                            X(16).
                                            X(16).
                                    PICTURE X(16) .
                                    PICTURE X(16).
                                    PICTURE 9(8).


                                    PICTURE 9(8).
                                    PICTURE 9(8).


                                    PICTURE 9(5)V9(3) OCCURS 7.
                            nrrnot;
                        125

-------
            07  DATA-3-A
        05  SEG-3-ri.
            07  DATA-3-6
    03  SEG-4.
        05  SEb-f-A
            07  DATA-4-A
    03  SEG-5.
        05  ShIG-5-A.
            07  DATA-5-A
    03  SEG-fe.
        05  SFG-6-A.
            07  DATA-6-A
        05  SEG-6-8.
            07  DATA-6-8
    03  SEG-7.
        05  SEG-7-A.
            07  DATA-7-A
    03  SEG-8.
        05  SEG-8-A.
            07  DATA-fl-A
    03  FILLER
FD  STATION-MASTEK-F
    LABEL RECORD IS OMITTED
    RECORD CONTAINS ftO CHARACTERS
    DATA RECORD IS F-RECORD.
01  F-RECORD.
    03  F-REC
WORKING-STORAGE SECTION.
77  I               USAGE COMP
77  J               USAGE COMP
77  TALLY-REC       USAGE COMP
01  F-CONVERTED.
    03  F-SIGNED                PICTURE
01  TALLY-MESSAGE.
    03  TALLY-REC-X
    03  FILLER
                                    PICTURE 9(8) OCCURS 10.

                                    PICTURE 9(8) OCCURS h.

                            OCCURS 18.
                                    PICTURE S9(9)V9(3) OCCURS b.


                                    PICTURE 9(8) OCCURS 10.


                                    PICTURE 9(8) OCCURS 3.

                                    PICTURE V(5)V9(3) OCCURS 4.


                                    PICTURE 9(8) OCCURS 10.
                                    PICTURE 9(8) OCCURS 5.
                                    PICTURE X(92).
                                    PICTURE X(dO) .

                                    PICTURE 999.
                                    PICTURE 999.
                                    PICTURE 999 VALUE ZERO.
                                                 -9.999 OCCURS 5.
                                                   VALUh SPACES.
                                    PICTURE 2229.
                                    PICTURE X(E2)
            VALUE " RECORDS CONVERTED".
    03  FILLER                      PICTURE X(105)
PROCEDURE DIVISION.
AOOO-CONTROL SECTION.
A001.
    NOTE    THIS SECTION CONTROLS EXECUTION
            OF THE PROGRAM.
AOIO-START.
    OPEN OUTPUT PRINT-FILE.
    OPEN INPUT  STATION-MASTER-C.
    OPEN OUTPUT STATION-MASTER-F.
A030-READ-RECORD.
    READ STATION-MASTER-C
        AT END GO TO A090-STOP.
    PERFORM COOO-WRITE-CONVERTED.
    GO TO A030-READ-RECORD.
A090-STOP.
    CLOSE STATION-MASTER-C.
    CLOSE STATIUN-MASTER-F.
    MOVE TALLY-REC TO TALLY-REC-X.
    MOVE TALLY-MESSAGE TO PRINT-DATA.
    MOVE 1 TO CARRIAGE-CONTROL.
    WRITE PRINT-LINE.
    CLOSE PRINT-FILE.
    STOP RUN.
COOO-WRITE-CONVERTED SECTION.
COOl.
    NOTE    THIS SECTION CONVERTS THE COBOL RECORD TO FORTRAN
            COMPATIBLE BY BREAKING THE RECORD INTO
            «n THARACTFR BCD wFrr»Rn«;.
                                     126

-------
       CO10-CONVERT-RECORD.
           MOVE SEG-1 TU F-RECORD.
           WRITE F-RtCORD.
           WOVE SEG-? TO F-RECOHO.
           WRITE F-KECORD.
           PERFORM MOVE-SE6-3-A
               VARYING I FROM 1 BY 1 UNTIL I GREATER THAN 6.
           MOVE SEG-3-B TO F-RECORD.
           WRITE F-RECORD.
           PERFORM MOVE-SEG-4-A
               VARYING I FROM 1 BY 1 UNTIL I GREATER THAN IB.
           MOVE SEG-5 TO F-RECORD.
           WRITE F-KECORD.
           MOVE SEG-b TO F-RECORD.
           WRITE F-RECORD.
           MOVE SEG-7 TO F-RECORO.
           WRITE F-RECORO.
           MOVE SEG-8 TO F-RECORD.
           WRITE F-RECORD.
           ADD i TO TALLY-REC.
           GO TO C090-EXIT.
       MOVE-SEG-3-A.
           MOVE SEG-3-A (I) TO F-RECORD.
           WRITE F-RECORD.
       MOVE-SEG-4-A.
           PERFORM MOVE-DATA-4-A
               VARYING J FROM 1 BY 1 UNTIL J GREATER THAN 5.
           MOVE F-CONVERTED TO F-RECORD.
           WRITE F-RECORD.
       MOVE-DATA-4-A.
           MOVE DATA-4-A (I J) TO F-SIGNED (J).
       C090-EXIT.
           EXIT.
&
      PROGRAM CNVRT2 (TAPEX.STNFIL,OUTPUT
     $»        TAPE10 = TAPEX.TAPE11 = STNFIL»TAPEM=OUTPUT)
C
C  PROGRAM TO CONVERT THE INTERMEDIATE FORTRAN TAPE (BCD)
C    TO THE FINAL REDUCED STATION FILE (BINARY)
C
C  STATION FILE RECORD
C    WORD      DESCRIPTION
C                             DATA IS INTEGER UNLESS OTHERWISE SPECIFIED
C       1  STATION NUMBER
C     2-3  STATE NAME         ALPHA
C     4-5  COUNTY NAME        ALPHA
C     6-7  STREAM NAME        ALPHA
C     8-9  LOCATION NAME      ALPHA
C      10  LATITUDE
C      11  LONGITUDE
C      12  TOTAL
C      13  TOTAL AREA
C      14  MAIN CHANNEL SLOPE          FL. PT.
C      15  RAIN INTENSITY              FL. PT.
C      16  RAIN EROblON INDEX          FL. PT.
C      17  SLOPE PCT CLASS             FL. PT.
C      18 EROSION CLASS                FL. PT.
C      19  SOIL INFILTRATION           FL. PT.
C      20  SOIL INDEX                  FL. PT.
C   21-52  SOIL CLASS AREA TABLE  (4X8) (4 CLASSES. 8 TYPES)
C   53-84  SOIL RESOURCE GROUPS   (4X8) (4 CLASSES* 8 TYPES)
C      85  DATA YEAR (1969)
C  86-175  WATER QUALITY TABLE (18X5) (18 PARAMETERS* 5 PERIODS)
C              18 PARAMETERS = SEE BELOW
C               5 PERIODS = ANNUAL* SPRING, SUMMER. FALL* WINTER
r          WflTFR OIIAI TTY PARAMFTFR*;    Fl OftTTMfi PT.

                              127

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c















176
177
178
179
180
Ifll
182
183
184
185
186
187
188
189-192
193
194
195
196
197
198
199
200
201
202
203-207
208-210

1 TEMPEHATURt (F)
2 STREAMFLOW (CFS)
3 TURHIOITY (JU)
4 DO (MGL)
5 HOD (MGL)
6 PHOSPHORUS (MGL)
7 TOTAL COLIFORM (100ML)
8 SUSPENDED SOLIDS (MftL)
9 DISSOLVED SOLIDS (TAF)
10 NITRATE (MGL)
11 HERHICIDE (TY)
12 INSECTICIDE (TY)
13 FUNGICIDE (TY)
14-18 NOT USED

LAND AREA
LAND IN FARMb
OTHER LAND
TOTAL CROP AHFA
CROPLAND HARVESTED
PASTURE * GRAZING
OTHER CROPLAND
IRRIGATED
OTHER FARMLAND
ROW CROPS
SMALL GRAIN
HAY
WOODLANDS
C-FACTORS (4) (4 CROP GROUPS) FL. PT.
CATTLE
HOGS
SHEEP
CHICKENS
NITROGEN
PHOSPHATE
POTASH
HERBICIDE
INSECTICIDE
FUNGICIDE
RAINFALL ( ANNUAL » SPRING. SUMMER .FALL » WINTER)
NOT USED

DIMENSION IBUF<210)» RAIN(5)
EQUIVALENCE (IBUF(203)» RAIN(D)
C


LUBCO = 10
LUBIN = 11
LUPRNT = 61
NRECS = 0
C
C  READ THE 80-CHARACTER BCD RECORDS OF THE NEXT LOGICAL RECORD
C
  100 CONTINUE
      READ (LUBCD.6001)  (IBUF(I),1 = 1 * 10)
 6001 FORMAT  (I8»4(A10«A6)* 18)
      IF (EOF(LUBCD)) 500*105
  105 READ (LUBCD.6002)  (IPUF (I) , 1 = 11,19)'
 6002 FORMAT  (2I8»7F8.3)
      L = 19
      DO 150 J = 1*6
      K = L + 1
      L = K * 9
      READ (LUBCD»6003)  (IBUF(I),I=K»L)
 6003 FORMAT  (1018)
  150 CONTINUE
      RFAD M iiRrn.Ann^i  i THIIF i n .T=*I

                                  128

-------
      L = 85
      DO 200 J = 1,1*
      K = L + 1
      L = K + 4-
      READ  (LUBCD,6004)
 6004 FORMAT (5F14.3)
  200 CONTINUE
      READ  (LUBCD,6003)
      READ  (LUBCD,600b)
 6005 FORMAT (3I8,4F8.3)
      READ  (LUBCD,fr003)
      READ  
-------
5.  SORT FILE INTO 13 SUBBASIN FILES
                130

-------
BARRY»CM10000,T20.
ACCOUNT,M350001»MIRWEST,3807C.
GET*TAPE10=BDS3807.
REWIND,BDS3807.
RFL«40000.
RUN23CS)
RFL»70000.
LGO.
SAVE»TAPE11=BES3807.
GET»BES3807.
REWIND,BES3807.
COPYSBF(BES3807,OUTPUT)
GET COST(LIBRARY)
RFLflOOOO.
COST.
EXIT.
GET COST(LIBRARY)
COST.
&
      PROGRAM SORT
     1
     2
             (TAPE10»TAPE11»TAPE31»TAPE32,TAPE33,TAPE34,TAPE35,
              TAPE36,TAPE37»TAPE38,TAPE39,TAPE40,TAPE41,TAPE42,
              TAPE43, INPUT,OUTPUT,TAPE12=INPUT»TAPE13=OUTPUT)
C
C	
C
   10

C
C	
C
   12
   13
   14
C
C	
C
DIMENSION ISUBS(233)»ARAY<80)

READ SUB-BASIN COOES

READ (12»5) ISUBS
FORMAT (2513,5X)
REWIND 12
DO 10 1=31,43
REWIND I
CONTINUE
IREC=0

READ NEXT BASIN CAHD

IREC = IREC + 1
READ (10,13) ARAY
IF (EOF,10) 1000,14
FORMAT (20A4)
ISUB=ISUBS(IREC)

WRITE IN PROPER SUB-BASIN FILE
    < WRITE  (ISUB.13) ARAY
      GO TO  12
 1000 DO 1001 1=31,43
      REWIND  I
 1001 CONTINUE
      DO 1100 1=31,43
 1050 READ  (1,13) ARAY
      IF (EOF»I)  1100,1075
 1075 WRITE  (11,13)  ARAY
      GO TO  1050
 1100 CONTINUE
      STOP
      END
&
&
                          131

-------
6.  ADD ANNUAL RAINFALL TO FILE

-------
      PROGRAM CNVKT3  (OLDFlL.NF.wFIL .OUTPUT
     5»        TAPE10=OLOFIL.TAP£11=NEWFIL,TAPE61=OUTPUT)
C
C  PROGRAM TO CONVERT KAINFALL DATA ON THE STN. FILE
C
      DIMENSION IBUF(210), REAO(5)
C
      LUOLD  = 10
      LUNEW  = 11
      LUPRNT = 61
      NRECS  = 0
C
C  GET OLD RECORD
C
  100 BUFFER IN (LUOLD.l)  (IHUF(1),IBUF(210))
      IF  (UNIT(LUOLD)) 150,600,125
c        * READ PARITY *
  125 STOP 2010
C
C  CONVERT TO NEW RECORD
C
  150 ANNUAL = 0.0
      DO 175 I = If4
      J = 207 - I
      K = 6 - I
      RAIN(K) = 0.001 * FLOAT( IBUF(J) )
  175 ANNUAL = ANNUAL + RAIN(K)
      RAIN(I) = ANNUAL
C
C  WRITE THE NEW RECORD
C
      BUFFER OUT (LUNEw,!) (IBUF(1),IBUF(210))
      IF  (UNIT(LUNEtv) ) 200,225,250
  200 NRECS = NRECS + 1
      GO TO 100
C        * EOF ON WHITE »
  225 STOP 1011
C        * WRITE PARITY *
  250 STOP 2011
C
C  CONVERSION COMPLETE
C
  500 ENDFILE LUNEw
      WRITE (LUPRNT,6101)  NRECS
 6101 FORMAT (1H1«I4«1BH RECORDS  CONVERTED  )
      STOP
      END
                               133

-------
7.  SAMPLE OF SPSS RUNS
            134

-------
BARRY»T150,CM10000.
ACCOUNT,M350001,MIRrtEST;3807C.
GET,BES3807.
REWIND,BES3807.
COMMON(MRISPSS)
REWIND,MRISPSS.
RFL,120000.
MRISPSS.
GET»COST(LIBRARY)
RFL,10000.
COST.
MRISPSS
HUN NAME
VARIABLE LIST
SUBFILE LIST
INPUT MEDIUM
NO. OF CASES
INPUT FORMAT
VAR LABELS
COMPUTE
VAR LABELS
MISSING VALUES
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
VAR LABELS

MISSING VALUES
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
COMPUTE
rnMPIITF
                                 CORRELATION, REGRESSION BY SUB-BASIN
3807C
BASIC STATISTICS,
VAR001 TO VAH036
S31,S32,S33,S34,S35,S36,S37,S38,S39,S40,S41,S42,S43
BES3807
28,41,20,27,32,18,9,9,5,4,5,11,24
FIXED(F8.0»6F12.3/4F12.3/10F8.0/6F8.0,F8.3/4F8.3,4F8.0)
VAR001»STATION/VAK002,STREAMFLOW-CPS/
VAR003,TUPBIDITY-JU/VAR004,DO-MGL/
VAR005,BOD-MGL/VAR006,PHOSPHOROUS-MGL/
VARO07,TOTAL COLIFORM-1—1OOML/VAR008,
TOTAL COLIFORM-2—100ML/ VAR009,FILTERED RSD-MGL/
VAR010,DISSOLVED SOLIDS-TONS PER  DAY/
VAR011,NITRATE-MGL/VAR012,LAND AREA/
VAR013,LAND IN FARMS/VAR014,TOTAL CROP AREA/
VAR015,CROPLAND HARVESTED/VAR016,PASTURt AND GRAZING/
VAR017,OTHER CROPLAND/VAR018,IRRIGATED LAND/
VARO19,OTHER FARMLAND/VAR020,CATTLE/VARO21,HOGS/
VAR022»SHEEP/VAR023,CHICKENS/VAROP4,NITROGEN/
VAR025,PHOSPHATE/VAR026,HERBICIDES/VAR027,INSECT I SIDES/
VAR028,ANNUAL RAINFALL/VAR029,RAIN INTENSITY/
VAR030,SLOPE PCT CLASS/VAR031,EROSION CLASS/
VAR032,SOIL INFILTRATION/VAR033,ROW CROPS/
VAR034,SMALL GRAIN/VAR035,HAY/VAR036,WOODLANDS
VAR037=VAR007 + VAR008
VAR037,COLIFORM TOTALS
VAR001 TO VAR037(0)
VAR038=VAR020 + VAR021
VAR039=(1.0*VAR020)+(.25*VAR021)+(
VAR048=(1.0*VAR020)+(,09*VAR021)+(
VAR047=(1.0*VAH020)+(.167*VAR021)+
VAR040 = VAR024 + VAR025
VAR038,CATTLE AND HOGS/VAR039,ALL LIVESTOCK/
VAR040,NITROGEN AND PHOSPHATE
VAR038 TO VAR040(0)
       ={VAR002 * VAR004) * 5
                  VAR005) * 5
                  VAR006) * 5
                  VAR011) * 5
                                                  ,15*VAR022)+(
                                                  ,06*VAR022)*(
                                                  [.067*VAR022)
                                               .015*VAR023)
                                               ,006*VAR023)
                                               •M.0067*VAR023)
VAR041
VAR042
VAR043
VAR044
VAR010
VAR038
       =(VAR002
       =(VAR002
       =(VAR002
       = VAR010
       =   (VAR038 / VAR012)
VAR047=VAR047/VAP012
VAR048=VAR048/VAR012
                                             ,39
                                             ,39
                                             ,39
                                             ,39
VAR039 =
VAR024 =
VAR040 =
VAR020 =
VAPfl?1 =
(VAR039 / VAR012)
(VAR024 / VAR012)
(VAR040 / VAR012)
(VAR020 / VAR012)
f W A B n ? 1 / U A B n 1 9 t
                                  135

-------
COMPUTE        VAW027=VAN027  /  V«rt012
COMPUTE        VAROlh  =  VARU16  /  VAP012
COMPUTE        VAK02S  =.   (VAKOP'j / VAR012)
COMPUTE        VAP01H  =    (VAHOlH / VAR012)
COMPUTE        VAH014=VAK014/VAK012
VAR LABELS     VAr«041 ,0(J-LH-PEW-GAY/VAK042»BOD-LB-PER-L)AY
               VAR043»P-L8-PEH-OAY/VAR044«N03-LB-PEk-L)AY
HISSING VALUES VARIKI  TO VAR044> + 
COMPUTE        VAR078=LN(VAK028)
COMPUTE        VAR079=LN(VAP029)
COMPUTE        VAROBO=LMVAft030>
COMPUTE        VAR082=LN(VAR032)
COMPUTE        VAR09b=LN(VA*045)
MISSING VALUES VAR050 » VAP053* VAH064, VARO 70 » VAR071 < VAR07b. VAR07H, VAR080 . VAR082,
 VAR  LABELS '     VAR050.TURB-PEK-SQ-MI/VAROb3,TURB-Pt"R-SQ-MI/VAR064»LOG-CROPLANO
                PER SQ MI/VAP070,LOG-CATTLE-PER-SQ-MILE/VAR071,LOG-HOGS-PtR-SQ-MI
                LE/VAR075»LOG-PHOS-PER-S«-MI/VAR078.LOG-RAINFALL/VAR079,LOG-RAINF
                ALL INTENSITY/VAR080.LOG-SLOPE-PCT-CLASS/VAR082. LOG-SOIL INFILTRA
                TRATION RATE/VAP095.LOG-C FACTOR/
 COMPUTE         VARO
-------
8.  USE FILE TO GET TABLES SHOWING COEFFICIENT  OF VARIATION  BETWEEN
        PREDICTED AND OBSERVED VALUES OF DEPENDENT VARIABLES
                                 137

-------
BARRY »CM60000»T1 00.
ACCOUNT, M350001,MIRWEST,3807C.
GET»TAPE3=BES3807.
REWIND»TAPE3.
FTN,R=2.
MAP.
LGO.
GET,COST(LIBRARY)
RFL»10000.
COST.
EXIT.
GET,COST(LIBRARY)
RFL»10000.
COST.
&
      PROGRAM  COMPARE  ,
      1) »VALS(17) ;ICOUNT(3»14) » SIGN (3, 11) ,BASIN(14) , TEMP < 13)
      DIMENSION DEPVARO)
      DATA  PLUS/1H+/
            PMINUS/1H-/
            I SUBS/28,41,20,27,32,18,9,9,5,4,5,11,24/
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
      DATA
            BASIN(l)  /9H   31     /
            BASIN(2)  /9H   32     /
            BASINO)  /9H   33     /
            BASIN14)  /9H   34     /
            RASIN(5)  /9H   35     /
            BASIN16)  /9H   36     /
            BASIN<7)  /9H   37     /
            BASIN(8)  /9H   38     /
            BASIN(9)  /9H   39     /
            BASIN(10)/9H   40     /
            BASINUD/9H   41     /
            BASIN(12)/9H   42     /
            BASIN(13)/9H   43     /
            BASIN(14)/9H   ALL   /
            DEPVAR(1)/10HTUR-X-FLOW/
            DEPVAR(2)/10HBOD-PPD/M2/
            DEPVAR(3)/10HNIT-PPD/M2/
C
C ----
C
   29
    1
   11
    2
       INITIALIZE ARRAYS

       DO 2 1=1,14
       DO 29 J=l,3
       ICOUNT(J,I)=0
       YBARdf J)=0.0
       COV(I,J)=0.0
       CONTINUE
       DO 11 K=l,3
       DO 1 J=l,13
       AVGES(K,I»J)=0.0 x
       CONTINUE
       CONTINUE
       CONTINUE
  ---- READ COEFFICIENTS
      DO 3 1=1,11
      READ (1,101)
                    (COEFF(J,I) ,J=1,3)
   101
                                    138

-------
    3 CONTINUE
C
C ---- BEGIN LOOP TO READ SUB-HASIN VALUES
C
      DO 5 1=1,13
      J=ISUBS(I)
      DO 4 K=1,J
      READ (3,102) 
-------
     39)*AVGES(L.I»11)+COEFF(L,10)*AVGES(L»I,12)+COEFF(L»11)»AVGES(L,I»13
     43)
      IF(Y6AR(I.L) .EG. 0.0) GO TO 22
      COV(I,L) = ( (YBAK(I»L)-AVGFS(LiI,L) )/YBAR(I.L) )*100.0
      COV(I,L)=ABS(COV(I,L))
      GO TO 8
   22 COV(I,L)=100.0
    8 CONTINUE
    5 CONTINUE
C
C	 COMPUTE COEFFICIENT OF VARIATION  FOR ALL  SUHBASINS
C
      DO 209 N=l»3
      AVGES(N»14»N)=AVGtS(N,14,N)/ICOUNT(N,14)
      DO 207 L=4»13
      AVGES(Ntl4»L)=AVGES(N,14.L)/ICOUNT=COEFF*AVGES(L»14,10)+
     3COEFF(L.9)*AVGES(Ltl4,ll)+COEFF(L»10)»AVGES(L,14,12)+COEFF  ,A1,F8,
     3 »A1,F8.5»8H*LN(X9)  »A 1»F8.5»8H*LN(XI0))
      WRITE(2*105)  I.DEPVAR(I)»I»DEPVAR(I)
  105 FORMAT(/»9H WHEHE Y(»I1»22H)= OBSERVED VALUE OF   .AID
     It      /»12H       YBAR(»I1»21H)=PREDICTED  VALUE  OF  «A10
     2»      /»30H       X(1)=CROPLAND/SQ MILE
     3,      /»32H       X(2)=PASTURE  LAND/SQ MILE
     4*      /»37H       X(3)=IRRIGATED. LAND/SQ MILE
     5t      /,30H       X(4)=CATTLE/SQ MILE
     6»      /.30H       X(5)=HOGS/SO  MILE
     7»      /,30H       X(6)=NITROGEN/SQ MILE
     8t      /t32H       X(7)=PHOSPHATE/SQ MILE
     9»      /.30H       X(8)=RAINFALL
     A*      /.30H       X(9)=SLOPE-PCT-CLASS
     Bt      /»30H       X(10)=C FACTOR
     Ct///55H COEFFICIENT  OF VARIATIONS ABS(YBAR  - Y(D)    *   100.0
     Dt/31XtllH	
     Et/»3'4X»4HYBAR        )
C
C	 WRTTF NFW PAfiF  HFADTNfi
                                 140

-------
      WRITE<2*107>
      WRITE<2»106) I.I
  107 FORMATdH »117X»15HNO OF  COEFF OF)
  106 FORMATU04H BASIN NO.  X(l)     X(2)     X{3)
     1  X(6)     X(7)     X(8)     X(9)    X(10)
     2I1»17H)  OBS   VAR.PCT.)
                                                        X(4>
                                                              X(5)
                                                              YBAR(*
c ---- WRITE COEFFICIRNTS
C
      DO 41 J=l»14
      WRITE (2> 108) BASIN(J) , ( AVGES ( I , J»K) »K=4,13) , AVGES ( I » J* I ) »YBAR(J»D
     l.ICOUNT(I»J) »COV(J,I)
  108 FOPMAT(/»A9»12F9.5,I4,F10.2)
      CONTINUE
      CONTINUE
      STOP
41
 9
   END
 -8.63206
-4.47193
  0.0
0.0
2.53051
.23789
0.0
.72239
-1.10095
1.2993
0.43524
&
#
            -.80188
            -2.31428
            0.0
            0.03081
            -.25308
            -.18659
            0.0
            0.0
            2.58872
            0.0
            1.74135
                              -11.17613
                              -1.88129
                              -.21286
                              -.0791
                              1.82359
                              -.19982
                              .62432
                              0.0
                              1.17815
                              0.0
                              .62061
                                  141

-------
9.  PRINT TABLES OF INDEPENDENT VERSUS DEPENDENT VARIABLES GIVING
               CORRELATION AND REGRESSION ANALYSIS
                                142

-------
SUMMARY,Tl00,CM10000.
ACCOUNT,M3504l5,MIKivfc-ST,3807C.
RFL»50000.
FTN.
COMMON.RJP3807.
REWIND»RJP3807.
REWIND,TBL3807.
COMMON,TBL3807.
RFL»70000.
LGO(RJP3807,,TBL3807)
REWIND,TBL3807.
COPYBF(TBL3807,OUTPUT)
GET»COST(LIBRARY)
RFL»10000.
COST.
EXIT.
REWIND,TBL3807.
COPYBF(TBL3807,OUTPUT)
6ET,COST(LIBRARY)
RFL,10000.
COST.
&
      PROGRAM SUMMARY  (LSTFIL,"INPUT,OUTPUT
     $,                 TAPE20=LSTFIL,TAPE60=INPUT.TAPE61=OUTPUT>
C
C  PROGRAM TO READ A COPY OF  THE OUTPUT FILE FROU A SPECIAL SPSS
C    RUN WHICH INCLUDES MARGINALS, PEARSON CORRELATIONS. AND REGRESSIONS
C    AGAINST THE SUB-BASIN SEQUENCED STATION DATA FILE
C
C
C  DEFINITIONS OF VARIABLES
C
C    LINE   = LINE IMAGE BUFFER FOR ONE LINE FROM THE LIST FILL
C    IVPTR  = VARIABLE  POINTER TABLE, ENTRY GIVES INDEX OF VARIABLE
C                IN  REDUCED SET FOR THIS PROGRAM
C    IVNAM  = 10-CHARACTER NAMES OF VARIABLES  IN REDUCED SET
C    IMARG  = MARGINALS STATISTICS TABLE   (14X13X3)
C                14  SUB-BASINS  (31-43, ALL)
C                13  VARIABLES   (33-36,10,20-21,30-31,24-25,32,18)
C                 3  STATISTICS  (MEAN, NO.  OBS,  STD DEV)
C    ICORR  = CORRELATIONS STATISTICS TABLE   (14X40X3)
C                14  SUB-BASINS   (31-43, ALL)
C                40  CORRELATIONS  (5 DEPENDENT  X H INDEPENDENT)
C                 3  STATISTICS    (COEFF.,  SIG,  NO. PAIRS)
C    IREGR  = REGRESSIONS STATISTICS TABLE   (14X40X6)
C                14  SUB-BASINS   (31-43, ALL)
C                40  REGRESSIONS  (5 DEPENDENT  X  8  INDEPENDENT)
C                 3  STATISTICS   (A, B, F)  (A  TAKES 2 LOCATIONS)
C                      AND TWO DEGREES OF FREEDOM
C
      COMMON /Bl/ LINE(14), LUNLST, LUSLST, LXSUH, IXSUH
C
      DIMENSION   IVPTK(48>,  KEYSK4), IMARG (14, 16, 3) ,  JMARG(3)
     $,           ICORR(14,48,3),  IREGR(14,48,6)
     $,           IVNAMU4).  IHEAD(2)
      DIMENSION  IVPTSU8)
C
      EQUIVALENCE (LINE(l), IWORD1)
C
      DATA  I IINI ^T/?0/. I UWrRO/hn/.  IIINPPT/A1/
                             143

-------
      DATA IVPTR/
     $
      DATA IVPTS /
     $
     $
     $
     S

      DATA ( (
     $3H
                   0*0,b,0,0.0,0,0,0*4,
                   0.0.0.14.0,0,0,lb,0*7*
                   M . 0 . 0 . 12 . 13 » 0 » 0 . 16 . 0 . 0 »
                   0.0.0.0,0,0,0,0*9*0*
                   b.1,2,3,0,0,10,11     /
                  0.0.5.0.0.0,0*0*0.4.
                  0.0.0.12.0.0.0.13.0*7*
                  H.0.0.10,11.0.0.14,0.0.
                  0.0*0*0*0*0*0*0*9*0*
                  6,1*2,3,0,0,9*9        /

                   ;i*J,K),K=l*b)*I = 1.14)»J=l»4ii)
C
C
C
C
C
C
C
C
C
C
C
      DATA  KEYS1/ 1OHVA&IAHLE  .10HMEAN
     S.            10HVALID   0*  /
                                              .10HSTD DEV
     DATA  IMARG/672*1H /*  ICORR/2016*1H /
     DATA IVNAM/  1 OH*Ob-PPD/M2» 1 OHP-PPD/M2   , 10HN03-PPD/M2
    $.            10HCS-TPU/M2  .10HTUR/M2     , 1 OHDO-PPO/M2
    $»            10HCAT/SU MI  .10HH06S/SU ^1 , 10ML VSTK/SUMI
    $*            10HNITRO/SUMI.10HPHOS/SQ MI * 10HCROP/SQ MI
    $.            10HIRRI/SQ MI,10HRAINFALL       /

                      #««#««#«•»«•«
                        MAH6INALS SECTION
  TOP OF LOOP BY  STATISTIC
  150
     LXSUB =  0
     IXSUB =  0
     DO 300 IXSTAT =
1.4
  GET NEXT LINE  IMAGt
 200 CALL GETLIN
     IF  (IXSUB - LXSUH)
 205 LXSUB =  IXSUB
                         600.210.205
C
C
  BRANCH ACCORDING  TO LINE  CONThNT
 210  IF  (IWORD1.NE.KEYSKIXSTAT)) GO  TO  200
      GO TO  (220,240,260.2flO)»IXSTAT

I  DECODE VARIABLE NOMbfcH
 220  DECODE  <  17,5101.LINE)  JXVAK
5101  FORMAT  (14X.I3)
      KXVAP =  IVPTR(JXVAP-)

  CHECK  IF VARIABLE IS  NEEDED
      IF  (KXVAR.LE.O)  GO TO  400
      GO  TO 300
   DECODE MEAN
  240 DECODE  ( 22,5102,LINE)
 5102 FORMAT  <12X,A10)
      GO TO 300
                             IMARG(IXSUB,KXVAP,1)
C
C
  DECODE STD DEV
 260 DECODE  (  22,5102,LINE)  IMAPG
-------
C  END OF STATISTICS LOOP
  300     CONTINUE
C
C  CONTINUE SCAN OF MAH
  400 GO TO 150
C
C  OVERRIDE SUB-BASIN STATISTICS
  600 KXSUB = 0
  620 READ  
      LUSCRD = IXEOF(LUNCRO)
      IF (LUSCRO.EQ.l) GO  TO 600
      IF (JXSUB.Ed.999) GO  TO 800
      IF   = JMARti(l)
      IMARG(KXSUB«KXVAH.?)  = JMAwGO)
      IMARG(KXSUBtKXVAK,3)  = J«AHG(2)
      GO TO 620
C
C                      »*»****»»#»•»
C                   -»«««•* CORRELATION SECTION »*««»
C                      »»*»*»»»#*#»
C
C  TOP OF LOOP FOR  RUWS
  800 LXSUB = 1XSUB
  810   LXCOMB = 0
C
  815   IXCOMH = LXCOMH *  1
        LXCOMB = IXCOMh +  b
      IF{LXCOMB.GT.b.ANU.LXCOMH.LT.14)LXCOMH=H
      IF(LXCOMB.GT.P4.AND.LXCOMB.LT.30)LXCOMB=24
      IF (LXCOMB.GT.4fi)LXCOMH = 48
C
C  LOOK FOR FIRST LINE OF  NEXT PO*
  850 CALL  GETLIN
      IF (IXSUB - LXSUH) 1*00»B70»860
  860 LXSUB = IXSUB
  870 DECODE  (4.5201»LINE)  IUATA
 5201 FORMAT  (A4)
      IF (IDATA.NE.4HVAKO)  GO TO  H50
C
C  DECODE COEFFICIENTS
      DECODE  (120»5202*LINE)  (ICORR(IXSUB»1,1)»I = IXCOMB,LXCOMb)
 5202 FORMAT  (6(1 OX,A7,3X))
C
C  READ NUMBERS OF  PAlHb
      READ  (LUNLST.5203)  (ICORR(IXSU6.I.3)»1=1XCOMb,LXCOMR)
 5203 FORMAT  (6(12X»A5»3X))
C
C  READ SIGNIFICANCE LEVELS
      READ  (LUNLSTf5204)  (ICORR(IXSUB,I»2>»I=IXCOMB,LXCOMB)
 5204 FORMAT  (6(13X.A5.2X))
      DO 1050 I =  IXCOMB»LXCOMB
      IF  (ICORR(IXSU8,I,2).EQ.SH  ****)  ICORR(IXSUB»I.1)  = 7H
 1050 CONTINUE
C
C  END OF R0« LOOP
      IF  (LXCOMB .Gt. 4B)  GO  TO HlO
      GO TO 815
C
C
f.                      «.«*•»«•#«*»•*»»

                            145

-------
C                  *»»**  HFbKtSSIONS  SECTION **«»*
C                       «**tttt**«»»*e
C
C  TOP OF REGRESSION  SCflN LOOP
 1400 LXSUH =  IXSUrt
C
C  LOOK FOR DEPENDENT VAKlAHLE
 1450 CALL GETLIN
      IF  (LUSLST.KO.l)  fan TO  if000
      IF  (IXSUrt  -  LXSUh)  14t-0,1470,1460
 1460 LXSUb =  IXSUb
 1470 IF  (iWORDl.NE.lOriUEf'ENOEMT  >  GO TO 14SU
      DECODE  (30»53US»l.Ir.'E)  IXDfcP
 5305 FOHMAT  <27X,I3)
C
C     LOOK FOR LINE  «ITH  DEGREE  OF  FREEDOM
C
 1475 CALL GETLIN
      IF  (IV.ORD1.NE.1UHK  SUUARF   )  GO TO I47b
      DECODE(75.5301»LINF)  IDF1
 5301 FOWMAT(72X.A3)
      RF.AD(LUNLST»53U6)  IDF?
 5306 FORMAT(73X,A3)
C
C  LOOK FOR LINE «IT(- VARIABLE  INUEX, H, AND F
 1480 CALL 6ETLIN
      DECODE  ( 4»530?»LINE)  IOATA
 5302 FORMAT  (A4)
      IF  (IDATA.NE.4HV£H(0)  GO TO  1480
C
C  GET VARIABLE  INDEX, H, F
      DECODE  (70,5303,LINE)  IXlNU,  IUATA1, IOATA2
 5303 FOKMAT  (4X,12,10X,A10,34X,A10)
      IF  (IDATA2.E0.10H         K)  IOATA2 = 7H   ****
      IXDEP =  IVPTS(IXD£P)
      IXIND =  IVPTS(IXIND)
      IXCOM8  = H*(IXDEP - 1)  +  (IXIND - t>)
      IREGR(IXSUB»IXCOMH,3)=  IOATA1
      IREGR(IXSUB,IXCOM(3,4)=  IUATA2
      IREGR(IXSUB,IXCOMh(,5)=  IDF1
      IMEGR(IXSUP,IXCOMtt,h)=  IDF2
C
C  GET CONSTANT  (/*)
      READ  (LUNLST,b304)  
-------
 2030 WRITE  (LUNPKT,6101)  ( iHLAu(I),1 = 1,2)
 6101 FORMAT  ( 1H1 ,38X * ^OHKEftUt'SSlUN  AND  CORRELATION  ANALYSIS  FOrf  ,2A10
     S//58H   DEPENDENT     TNDtPENOFNT    CORRELATION   SIGNIFICANT
     $, 6HNO. OF  ,lbX.4?H         STANDARD    RE'-RFSSION  EQUATION
     5>,10HR.EGRESSION  /4SH   VAKIABLE      VARIAHLt     COEFFICIENT
     *«S7HLEVEL                      MEAN      OHS.    DEVIATION
     1.30HY  = A * BX         SIGNIFICANCE
     $//6X»lHY»13X,lhX,33X,!HN,hX,2HOF,41X»1HA.4X,1HH,1IX,1HF)
C
C  TOP OF LOOP HY DEPtNUENT  VARIABLE
 2050 00 2300 IXDEP = 1,5
C
C  PRINT DEPENDENT VAHIuhLE  LINE
      WWITE  (LUNPRT,610?)  IVNAM(IXDEP),(IMARfi(IXsUb.IXUEP,I),1=1,3)
 610? FORMAT  (/2X,Al0,S4X»A 10»A8»2X,A 1O/)
C
C  TOP OF LOOP BY INDEPENDENT VANlABLE
      00 2200 IXIND = 7,14
C
C     DETERMINE WHICH DEPENDENT  VAPIABLES  IS USED
C
      IXIND2=IXINn
      IF(IXOEP.EW.l.ANU.IXlNP.GE.10)fiO  TO  2204
      IFdXDEP.EQ.?.OR.IXDEP.EQ.3)60 TO  2205
      IF(IXDEP.GE.4)  GO  TO  220(S
      GO TO  2210
 2205 IF(IXIND.E0.9)  IAIN'D2=10
      IF(IXIND.GE.IO)IXIND2=IXlND+2
      GO TO  2210
 2204 IXIND2=IXIND + i-
      GO TO  2210
 2206 IF(IXIND.GE.9>  I XIMD?=IXINO + 2
      GO TO  2210
C
C  PRINT INDEPENDENT  VARIABLE LINE
 2210 IXCOMB  = 8*
-------
C  BOTTOM OF SUB-BASINS LOOP
 2400 CONTINUE
      STOP
      END
      SUBROUTINE GETLIN
C
C  ROUTINE TO GET THE NEXT LINE  IMAGE  INTO  A  DECODABLE BUFFER
C
      COMMON /HI/ LINt<14), LUNLST, LUSLST. LXSUB,  IXSU6
C
      READ 
-------
10.  PROGRAM TO LIST TABLES OF STATIONS WITH VALID OBSERVATIONS
                  FOR VARIABLES BEING ANALYZED
                               149

-------
SUMMARY,!150,CM40000.
ACCOUNT,M350001,MIRWEST,3807C.
GET,TAPE10=BES3807.
FTN.
RFL»145000.
LGO.
GET,COST(LIBRARY)
RFL»10000.
COST.
EXIT.
GET.COST(LIRRARY)
RFL,10000.
COST.
&
      PROGRAM SBASIN  (TAPE31,TAPE3?»TAPE33»TAPE34,TAPE36,TAPE36,TAPE37
     S»               TAPE38,TAPE39,TAPE40»TAPE41,TAPE42,TAPE43,TAPE11
     $» TAPE51»TAPEb2»TAPElO.INPUT*OUTPUT ,TAPE60=INPUT,TAPEfcl=OUTPUT)
C
C  PROGRAM TO CREATE  A  SUB-BASIN GROUPED STATION DATA  FILE
C    FROM THE STATION-NUMBER  SEQUENCED  STATION  DATA  FILE
C
      DIMENSION ISUbS(233),VAR<87)
      DATA   LUNCRD/60/, LUNPRT/61/. LUNBAS/10/« LUNSTN/11/
C
C  POSITION .FILES
      REWIND LUNBAS
      DO 50  LUNSU6 =  31,43
      REWIND LUNSUB
   50 CONTINUE
C
C  READ SUB-BASIN CODES
      DO i 1=1,as
      ISUBS(I)=1
     1 CONTINUE
      DO 2 1=29,69
      ISU8S(I)=2
     2 CONTINUE
      DO 3 1=70,89
      ISUBS(I)=3
     3 CONTINUE
      DO 4 1=90,116
      ISUHS(I)=4
     4 CONTINUE
      DO 5 1=117,148
      ISUBS(I)=5
     5 CONTINUE
      DO 6 1=149,166
      ISUBS(I)=6
     6 CONTINUE
      DO 7 1=167,175
      ISUBS
-------
   ID CONTINUE
      DO 11 1=194.19H
      ISUi3S(I)=ll
   11 CONTINUE
      DO 12 1=199.209
      ISUBS(I)=12
   1? CONTINUE
      DO 13 1=210.233
      ISUHS(I)=13
   13 CONTINUE
C
C  INITIALIZATION
      IREC = 0
      WRITttLUNPRT,6.100)
 6100 FORMAT (1H1)
C
C  READ NEXT BASIN RECORD
  100 IREC = IREC +  1
      READ 
      60 TO 121
  122 VAR(K)=0.0
  121 CONTINUE
      VAR<37)=VAR<7) +VAH<8>
      VAR(3R)=VAR(20) + VAH(21)
      VAR(39)=VAR(20) +VAH(21) + VAR(22) * VAR123)
      VAR<40) =VAR<24) +  VAR(25)
      VAR(41)=VAR(2) » VAR(4) *  5.39 /VAR112)
      VAR(42) = VAR(2) *VAR(5) » 5.39/VAR(12)
              =VAR(2> * VAR(6) «5.39  /VAR(12)
              VAR(2)*VAH(11)  »5.39    /VAR(12)
      VAR(3)=VAR(3)/VAR(12)
      VAR(10)=VAR(10)/VAR(12)
C
C	 CONVERT TO  NATURAL LOGS
C
      DO 130 1=45.87
      11=1-43
      X=VAR(I1)
      VAR(I)=ALOG(VAMIl) )
      IF (X .EQ.  0.0) VAR(I)=0.0
  130 CONTINUE
C
C  DETERMINE SU8-BASIN CODE
  120 ISUB = ISUBS(IPEC)  + 30
      IF (ISUB.LE.O) GO TO 100
C
C  WRITE RECORD TO PROPER SUb-8ASIN FILE
      WRITE(ISUBt6201)VAP
 6201 FORMAT(6F12.3)
      GO TO 100
C
C  PRINT TABLES FOR  EACH  SUB-BASIN
  200 DO 300 ISUB =  31.43
      REWIND ISUB
      CALL STN(ISUB)
  300 CONTINUE
C
C  RECOMBINE SUB-BASIN  FILES INTO  ONE  STATION FILE
      REWIND LUNSTN
      DO 400 ISUB =  31.43
      PFWTND TSIIR
                            151

-------
  350 READ (ISUB.6201) VAR
      IF (EOF(ISUH)) 390,3*0
  380 WHITE (LUNSTN,6201) VAR
      GO TO 350
  390 CONTINUE
  400 CONTINUE
      REWIND LUNSTN
      CALL STN(LUNSTN)
      ENDFILE LUNSTN
C
C  NORMAL EXIT
      STOP
      END
      SUBROUTINE STMISUH)
      DIMENSION VAR(H7)
      DIMENSION IVNAM(l^)
      DIMENSION VAR2(H7,?33)
      DIMENSION IDEP(6),INOtP(8)
      INTEGER REC
      DATA IDEP/42.43.44.10,3.41/
      DATA INDEP /  20»21.39»24»25»14,18»28/
      DATA IVNAM/   10HBOD-PPD/M2.10HP-PPD/M2   ,1OHN03-PPD/MH
     S»             10HOS-TPD/M2  ,10HTUR/M2     .10HDO-PPD/M2
     $»             lOHCAT/Stl MI  .10HHOGS/SQ MI , 1OHLVSTK/SQMI
     $.             10HNITRO/SQMI,10HPHOS/SQ MI,10HCROP/SQ MI
     S,             10HIRRI/SO MI.10HRAINFALL       /
      IREAD = 0
C	 READ NEXT RECORD
C
   12 READ  (ISUfi»201> VAH
  201 FORMAT(6F12.3)
C
C	 CHECK FO END  OF FILE
C
   52 IF(EOFdSUB) ) 19.23
   23 IREAD = IREAD + 1
C
C	 IF NOT END OF FILE  STORE  DATA  IN  ARRAY
C
      DO 28 J=l«87
      VAR2(J»IREAD)=VAP(J)
   28 CONTINUE
      GO TO 12
C
C	 END  OF FILE  ENCOUNTERED.  BEGIN PHOCESSING
C
   19 REC=0
      IF  (REC  .EQ.   0  .AND.  ISUB .EO.  10)  WRITE(61.2501)
      IF  (REC  .EG.  0 .AND.  ISUB .NE. 10)  WRITE(61.250)  ISU8
 2501 FORMAT(1H1»55X.15HALL  SUB-bASINS  )
  250 FORMAT(1H1,55X.10HSUB-BASIN  .1*)
C
C	 BEGIN DO LOOP FOR DEPENDENT  VARIABLES
C
      DO  10 1=1.6
      INDX=IDEP(I)
C
C	 BEGIN DO LOOP FOR INDEPENDANT  VARIABLES
C
      DO 20 J=l.fl
      JNDX=INDEP(J)
      REC=REC * 5
C
C	 WRITE NEW  STATION HEADINGS
C
                            152

-------
      IF (REC .GT. 50) GO TO 38
      WRITE  (61,4125)
      WRITE  (61,4124) IVNAM
-------
11.  PROGRAM TO CREATE A FILE (BDS 3807) FROM TOTAL STATION
                             154

-------
BARRY1»T100,CM10000.
ACCOUNT,M350415,MIRWEST,3807C.
GET»STNFIL.
RFL»50000.
FTN.
LGO.
REPLACE»TAPE53=8DS3807.
REWIND,TA8LE1»TABLE2.
COPYBF.TABLE1,OUTPUT.
COPYBF,TABLE2»OUTPUT.
RFL,10000.
GET,COST(LI8RARY)
COST.
EXIT.
COPYBF,TABLE1,OUTPUT.
COPYBF,TABLE2,OUTPUT.
RFLvlOOOO.
GET»COST(LIBRARY)
COST.
&
      PROGRAM TABLES  (STNFIL»TABLE1,TABLE2»TAPE53
     S»               TAPE10=STNFIL»TAPE51=TARLE1,TAPES2=TABLE2)
C
C  PROGRAM TO BUILD WATER QUALITY AND LAND USE TABLES ON DISK
C    FROM THE STATION FILE  (REDUCED FORTRAN VERSION)
C  TABLE FILES MUST BE COPIED TO OUTPUT
C  STATION FILE RECORD
C    WORD      DESCRIPTION
C                             DATA IS INTEGER UNLESS OTHERWISE SPECIFIED
C       1  STATION NUMBER
C     2-3  STATE NAME         ALPHA
C     4-5  COUNTY NAME        ALPHA
C     6-7  STREAM NAME        ALPHA
C     8-9  LOCATION NAME      ALPHA
C      10  LATITUDE
C      11  LONGITUDE
C      12  TOTAL
C      13  TOTAL AREA
C      14  MAIN CHANNEL SLOPE          FL. PT.
C      15  RAIN INTENSITY              FL. PT.
C      16  RAIN EROSION INDEX          FL. PT.
C      17  SLOPE PCT  CLASS             FL. PT.
C      18 EROSION CLASS                FL. PT.
C      19  SOIL INFILTRATION           FL. PT.
C      20  SOIL INDEX                  FL. PT.
C   21-52  SOIL CLASS AREA TABLE (4X8) (4 CLASSES, 8 TYPES)
C   53-84  SOIL RESOURCE GROUPS  (4X8) (4 CLASSES, 8 TYPES)
C   '   85  DATA YEAR  (1969)
C  86-175  WATER QUALITY TABLE  (18X5) (18 PARAMETERS, 5 PERIODS)
C              18 PARAMETERS = SEE BELOW
C               5 PERIODS = ANNUAL, SPRING, SUMMER, FALL,  WINTER
C          WATER QUALITY PARAMETERS    FLOATING PT.
C             1  TEMPERATURE (F)
C             2  STREAMFLOW  (CFS>
C             3  TURBIDITY   (JU)
C             4  DO          (MGL)
C             5  BOD         (MGL)
C             6  PHOSPHORUS  (MGL)
C             7  TOTAL COLIFORM-1  (100ML)
r             R  TOTAI roi TFORM-?  (

                            155

-------
C             9  FILTERED HSi)  (MGL)
C            10  DISSOLVED SOLIDS  (TON/DAY)
C            11  NITWATK      (MGL)
C            12  HERhlCIDE    (TY)
C            13  INSECTICIDE  (TY)
C            14  FUNGICIDE    (TY)
C         15-18  NOT USED
C
C     176  LAND AWEA
C     177  LAND IN KARN-S
C     178  OTHER LAND
C     179  TOTAL CHOP AKFA
C     1RO  CROPLAND HAttVF.bTED
c     181  PASTURE + GRAZING
C     182  OTHER CROPLAND
C     183  IRRIGATED
C     184  OTHER FARMLAND
C     185  ROW CROPS
C     186  SMALL GRAIN
C     187  HAY
c     188  WOODLANDS
C 189-192  C-FACTOPS (4)  (4 CROP GROUPS)    FL. PT.
C     193  CATTLE
C     194  HOGS
C     195  SHEEP
C     196  CHICKENS
C     197  NITROGEN
C     198  PHOSPHATE
C     199  POTASH
C     200  HERBICIDE
C     201  INSECTICIDE
C     202  FUNGICIDE
C 203-207  RAINFALL  (ANNUAL *SPRING.SUMMtH»FALL. «INTFft)
C 208-210  NOT USED
C
      DIMENSION IRUF(210)
C
      LU10  = 10
      LU51  = 51
      LU52  = 52
      LINES = 51
      NPE   = 0
      REWIND LU10
C
C  GET THE NEXT RECORD
C
  100 BUFFER IN (LU10.1)  (IBUF(1)«I8UF(210))
      IF  (UNIT(LUIO)) 130.200.120
C       * PARITY ERROR *
  120 WRITE (LU51.5101)
      WRITE (LU52.5101)
 5101 FORMAT (28H **» PARITY ERROR. NEXT LINE  )
      NPE = NPE + 1
      IF  (NPE.GT.10) GO TO 200
      LINES = LINES + 1
C
C  4HITE  THE DATA
C
  130 IF  (LINES.NE.51)  GO TO  140
C       * WRITE HEADERS *
      WRITE (LU51.5102)
 5102 FORMAT (1H1»54X.23H1969 WATER QUALITY DATA //
     S»57H      STN    FLOh-CPS     TURfl-JU      DO-MGL     fiOD-MGL
     S.48H      TP-MGL       TCOLI1      TCOL2     RSD-MGL
     $.?4H      DS-TPD     N03-MbL      /)
      WPTTF (I HS?.S?n?)

                            156

-------
 5202 FORMAT <1H1,56X»18H19&9 LAND USE DATA //
     $»57H      STN    LAND    FARM    CROP    HAHV    GRAZ   OCROP
     S»5f»H   OFARM   CATTL    HOGS   CHICK       N     P04    HErtB
     $«16H   INSCT    WAIN    /)
      LINES = 0
C       * WRITE DATA LIME *
  140 WRITE (LU51,5103) IBUF<1),  (IBUF(I)11 = 87,96)
 5103 FORMAT (1H ,I8»10F12.3>
      WRITE (LU52,5203) IBUF(1),IBUF(176),IBUF(177),(IBUF(I),1=179,182)
     $»   IBUF(184),(IBUF(I),1=193,194),(IBUF(I),1=196,198)
     $,   (IBUF(I),1=200,201),IBUF(203)
 5203 FORMAT (1H ,18,1418,F8.3)
      WRITE(53,5204) IBUF(l),(IBUF(I),1=87,96),IBUF(176),IBUF(177),
     1          (IBUF(I),1 = 179,184),(IBUF(I),1 = 193,198),IBUF(200) ,
     2          IBUF(201),IBUF(203)
 5204 FORMAT(I8,6F12.3/ 4F12.3/1018/618,F8.3)
      LINES = LINES  + 1
      GO TO 100
C
C  TABLES ARE READY  TO  PKINT
  200 ENDFILE LU51
      ENDFILE LU52
      REWIND LU51
      REWIND LU52
      STOP
      END
&
                                   157

-------
12.   PROGRAM TO PRINT TABLES OF ALL STATION VALUES FOR
               VARIABLES TO BE ANALYZED
                           158

-------
TABLES»T100,CM10000.
ACCOUNT,M35n415»MIRWEST,3807C.
COMMON,STNFIL.
REWIND.STNFIL.
RFL»50000.
FTN.
RFL»77000.
MAP(P)
LGO.
REWIND,TAPE53.
SAVE,TAPE53=BES3807.
GET»COST(LIBRARY)
RFL,10000.
COST.
EXIT.
GET,COST(LIBRARY>
RFL»10000.
COST.
&
      PROGRAM S8ASIN (TAPE31,TAPE32,TAPE33,TAPE34,TAPE35,TAPE36,TAPt37
     $»               TAPE38,TAPE39,TAPE40,TAPE41,TAPE42,TAPE43,TAPE11
     S, TAPE51»TAPE53,STNFIL»INPUT,OUTPUT,TAPE60=INPUT,TAPE61=OUTPUT
     $,TAPE10=STNFIL)
C
C  PROGRAM TO CREATE A SUB-BASIN GROUPED STATION DATA FILE
C    FROM THE STATION-NUMBER SEQUENCED STATION DATA FILE
C
      DIMENSION  IBUF(210), ISUBS(233)
      DATA  LUNCRD/60/, LUNPRT/61/, LUNBAS/10/, LUNSTN/11/
C
C  POSITION FILES
      REWIND LUNBAS
      DO 50 LUNSUB = 31,43
      REWIND LUNSUB
   50 CONTINUE
C
C  READ SUB-BASIN CODES
      READ (LUNCRD,6001) ISUBS
 6001 FORMAT  <(25I3»5XM
C
C  INITIALIZATION
      IREC =  0
      WRITE(LUNPRT,6100)
 6100 FORMAT  (1H1)
C
C  READ NEXT  BASIN RECORD
  100 IREC =  IREC +  1
      BUFFER  IN  (LUNBAS,!)  (IBUF(1);IBUF(210))
      IF  (UNIT(LUNBAS)) 120,200,110
C        *READ PARITY*
C 110 STOP 1002
  110 CONTINUE
C
C  DETERMINE  SUB-BASIN CODE
  120 ISUB =  ISUBS(IREC)
      WRITE(LUNPRT,6101) IREC,IBUF(1),ISUB
      WRITE(LUNPRT,9302) IBUFU5)
 9302 FORMAT  (18)
 6101 FORMAT  ( 8H RECORD ,15,  9H STATION  ,I10,11H  SUB-BASIN  ,13  )
      TF  (TSIIR.I  F.O) fiO TO  100

                                 159

-------
c
C  WRITE RECORD TO HKOPF.P SUH-HAS1N FILF
      HUFFER OUT  USUH,I)  URUFU) .IHUF (210) >
      IF (UNITUSUWM  100.1hO»170
C        »WPITE EOF*
  160 STOP 1101
C        *WRITE PARITY*
  170 STOP 110?
C
C
C  PRINT TABLES FOR EACH SUH-6ASIN
C
  200 DO 300 ISUK = 31«*3
      REWIND ISUB
      CALL TABLES(ISUB)
  200 CONTINUE
C
C  HECOMBINE SUB-BASIN FILFS  INTO  ONE  STATION  FILt
      DO 400 ISUB = 31*43
      REWIND ISUH
  350 PUFFER IN  (ISUrl,!)  (IBUF (1) , IRUF ( 21 0 ) )
      IF (UNIT(ISUd))  370»400.360
c        «READ PARITY*
  360 STOP 2102
      BUFFER OUT(LUNSTN,!) (leUF(l) «Ir,UF( 210)1
      IF (UNIT(ISUB))350,400,360
  400 CONTINUE
      REWIND LUNSTN
      REWIND LUNSTN
      CALL TABLES(LUNSTN)
      E'NDFILE LUNSTN
c
C  NORMAL EXIT
      STOP
      END
      SUBROUTINE TABLES  (ISUB)
C
C  PROGRAM TO BUILD WATER  QUALITY  AND  LAND  USE TABLES  ON  DISK
C    FROM THE STATION  FILE  (REDUCED FORTRAN VERSION)
C       AND THEN PRINT  THEM  OUT
C
C  STATION FILE HECORD
C    WORD      DESCRIPTION
C                        DATA IS  INTEGER  (18)  UNLESS OTHERWISE  SPECIFIED
C        1  STATION NUMBER
C     2-3  STATE NAME          ALPHA  (A10.A6)
C     4-5  COUNTY NAME         ALPHA
C     6-7  STREAM NAME         ALPHA
C     8-9  LOCATION NAME       ALPHA
C       10  LATITUDE
C       11  LONGITUDE
C       12  TOTAL
C       13  TOTAL ARtA
C       14  MAIN CHANNEL  SLOPE          FL.  PT. (F8.3)
C       15  RAIN INTENSITY               FL.  PT.
C       16  RAIN EROSION  INDEX          FL.  PT.
C       17  SLOPE PCT CLASS              FL.  PT.
C       18 EROSION CLASS                 FL.  PT.
C       19  SOIL INFILTRATION            FL.  PT.
C       20  SOIL INDEX                   FL.  PT.
C   21-52  SOIL CLASS  AREA  TABLE  (4X8)  (4 CLASSES*  8 TYPES)
C   53-84  SOIL RESOURCE GROUPS   (4X8)  (4 CLASSES.  8 TYPES)
C       85  DATA YEAR  (1969)
C  86-175  WATER QUALITY TABLE  (18X5)  (18 PARAMETERS*  5 PERIODS)
C              18 PARAMETERS  =  SEE BELOU
r               c. pFtrinrx;  = ANMIIAI . QPWTWK. <;IIMMFR. FAI i .  UTNTFD
                               160

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

















176
177
178
179
180
181
182
183
184
185
186
187
188
189-192
193
194
195
196
197
198
199
200
201
202
203-207
208-210

WATER QUALITY PARAMETERS FLOATING PT. (F12.3)
1 TEMPEMTURF. (F)
2 STRFAMFLOw 
3 TURHIUITY (JU)
4 DO (MGL)
5 BOD (MGL)
6 PHOSPHORUS (MGL)
7 TOTAL COLIFOPM-1 (100ML)
8 TOTAL COLIFORM-2 (100ML)
9 FILTERED RSD (MfaL)
10 DISSOLVED SOLIDS (TON/DAY)
11 NITRATE (MGL)
12 HERBICIDE (TY)
13 INSECTICIDE (TY)
14 FUNGICIDE (TY)
15-18 NOT USEO

LAND AREA
LAND IN FARMS
OTHER LAND
TOTAL CROP AREA
CROPLAND HARVESTED
PASTURE «• GRAZING
OTHER CROPLAND
IRRIGATED
OTHER FARMLAND
ROW CROPS
SMALL GRAIN
HAY
WOODLANDS
C-FACTORS (4) (4 CROP GROUPS) FL. PT. (F8.3)
CATTLE
HOGS
SHEEP
CHICKENS
NITROGEN
PHOSPHATE
POTASH
HERBICIDE
INSECTICIDE
FUNGICIDE
RAINFALL (ANNUAL » SPRING.SUMMER .FALL. WlNTFk) FL.PT.
NOT USED

                                                              (Frt.3)
      DIMENSION 8UF(210)
      DIMENSION I8UF(210)
      DIMENSION TEMP1(8),TEMP2(9)
      EQUIVALENCE (IBUF(1),BUF(1))
      EQUIVALENCE (TEMPI(1),IBUF(89))
      EQUIVALENCE (TEMP2(1).IriUF<193))
      DATA  LU51/51/. LU52/52/.  LUNPRT/61/
C
      LU10  = ISUB
      NRECS = 0
      LINES = 51
      NPE   = 0
      REWIND LU10
      REWIND LU51
      REWIND LU52
C
C  GET THE NEXT RECORD
C
  100 BUFFER IN  (LU10.1)  (I8UF(1).IBUF(210))
      IF  (UNIT(LUIO))  130*200.120
C       * PARITY ERROR *
  1?0 WRTTF fl 1151 .51 01 )
                                 161

-------
      WHITE 
-------
C  TABLES ARE READY TO PRINT
  200 ENDFILE LU51
      ENOFILE LU52
      REWIND LUSI
      REWIND LU52
C
C  PRINT TABLES
  250 READ (LU51,6101)
 6101 FORMAT (132H
     4
     $
      IF 
      GO TO 300
C
C  PRINT RECORD COUNT
  400 WRITE (LUNPRT.6102) NWECS
 6102 FORMAT (1H1,I5»13N RECORDS READ )
C
C  NORMAL EXIT
      RETURN
      END
SAVE»TAPE11=STNFIL.
C  WRITE RECORD TO NEW BASIN FILE
  370 BUFFER OUT (LUNSTN,!) (IBUF<1),IBUF(210))
      IF (UNIT(LUNSTN))  350,380,390
C        »«RITE EOF*
  380 STOP 2001
C        *WRITE PARITY*
  390 STOP 2002
&
 41 39 35 31 31 41 41 41 41 40 40 40 40 43 43 43 43 43 42 42 43 43 43 43 43
 43 43 43 43 43 43 43 43 43 43 43 43 43 43 42 42 42 42 42 42 42 42 42 39 39
 39 39 38 38 38 38 38 38 36 38 38 37 37 37 37 37 37 37 37 37 36 36 36 36 36
 36 36 36 36 36 36 36 36 36 36 36 36 36 35 35 35 35 35 35 35 35 34 34 34 34
 34 34 34 34 33 33 34 34 34 34 34 34 34 33 33 33 33 34 34 34 34 33 33 33 34
 34 34 34 34 34 34 34 33 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35
 35 35 35 35 35 35 31 31 31 31 31 31 31 31 31 33 33 33 33 33 32 32 32 32 32
 32 32 33 33 33 33 33 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32
 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 32 31 31 31 31 31 31 31 31 31
 31 31 31 31 31 31 31 31
                                  163

-------
SELECTED WATER
RESOURCES ABSTRACTS

INPUT TRANSACTION FORM
                                        L Re.T-. No.
 w
 4. n">  Analysis of Nonpoint-Source Pollutants in  the
       Missouri Basin Region
          Dr.  A. D. McElroy,  Dr.  F.  Y.  Chiu,
          Dr.  A. Aleti
          Midwest Research  Institute
          425 Volker Boulevard
          Kansas City, Missouri   64110

12. spooring o uu.2t.on Environmental  Protection Agency
                                     4
IS. Sappk:::inu:y N- u
 5. Rcr-rt Tyj.it:

 6

 8. Penoiiinna Or"1 L-..non
   r.ccrl Nc
                                                         PE 1BA030 16 AFN 03
68-01-1863
l}   y;v- of r-.eijon
   Environmental Protection Agency report number, EPA-600/5-T5-OC&, March 1975
      A study was conducted  of  nonurban, nonpoint sources  of pollution
 in  the Missouri Basin utilizing  a unique, computer-based  data system.
 The Data Bank contains extensive information on land use,  soil,  climate,
 water quality, and other parameters  in the Missouri Basin.   The  current
 study was concerned with the adequacy of the Data Bank relative  to develop
 ment of nonpoint pollution  models.
      Evaluation of information in the Data Bank yielded detailed land use
 and water quality profiles  in  the basin.  The results show that  the Data
 Bank is a useful base to depict  the  basinwide relations between  various
 land uses and water quality.   Regression equations of these relationships
 were developed for individual  subbasins as well as for the entire basin.
 The regression equations, and  the accompanying statistical analysis of
 significance, indicated that the water quality parameters  N03, BOD, and
 turbidity correlate well with  land use on a basinwide basis.   Other
 parameters, namely phosphorus, dissolved oxygen, and dissolved solids,
 did not correlate well with land use.
lJ7u. Descriptors
|17b.
 ! 7c. COWRR Field & Group
H 18. Availability

1
I
B
• Abstractor



Midwest
39.

20.

Research
Security
(Report)
Securit.i
(P*ge)
Inst
Class,.

Class.

.
21.

22

N7o. of
Pages
Price

Send To:







WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON,
OX.
2O24O

In:.: lution
KRSiC 102 (REV. JUNE 1971)

-------
                      RESEARCH REPORTING  SERIES

Research reports of the Office of  Research  and Development,  Environmental
Protection Agency, have been  grouped  into five series.   These five broad
categories were established to facilitate further development and appli-
cation of environmental technology.   Elimination of traditional  grouping
was consciously planned to foster  technology transfer and a  maximum inter-
face in related fields.  The  five  series  are:

     1.  Environmental Health Effects Research
     2.  Environmental Protection  Technology
     3.  Ecological Research
     4.  Environmental Monitoring
     5.  Socioeconomic Environmental  Studies

This report has been  assigned to the  SOCIOECONOMIC ENVIRONMENTAL STUDIES
series.  This series  includes research on environmental management,
economic analysis, ecological impacts, comprehensive planning and fore-
casting and analysis  methodologies.   Included are tools for  determining
varying impacts of alternative policies,  analyses of environmental plan-
ning techniques at the regional, state and  local levels, and approaches
to measuring environmental quality perceptions,  as well as analysis.of
ecological and economic impacts  of environmental protection  measures.
Such topics as urban  form, industrial mix,  growth policies,  control and
organizational structure are  discussed in terms  of optimal environmental
performance.  These interdisciplinary studies and systems analyses are
presented in forms varying from  quantitative relational analyses to manage-
ment and policy-oriented reports.
                           EPA REVIEW NOTICE

This report  has  been reviewed by the Office of Research and Development,
EPA, and approved for publication.   Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency,  nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
           For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402

-------