'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
------- |