United States
Environmental Protection
Agency
Environmental Researc'
Laboratory
Athens GA 30613
Research and Development
vEPA
Planning Guide for
Evaluating Agricultural
Nonpoint Source
Water Quality Controls
-------
PLANNING GUIDE FOR EVALUATING AGRICULTURAL
NONPOINT SOURCE WATER QUALITY CONTROLS
by
Paul D. Robillard, Michael F. Walter, and Linda M. Bruckner
Cornell University
Ithaca, New York 14853
Project Members:
Cornell University
George L. Casler
Douglas A. Haith
Marian 0. Harris
Earl A. Lang
Raymond C. Loehr
John H. Martin, Jr.
Christine A. Shoemaker
Lawrence J. Tubbs
Washington State University
Scott G. Matulich
Brian L. McNeal
Vincent F. Obersinner
William Pietsch
Norman K. Whittlesey
Colorado State University
Wynn R. Walker
Grant Number R804925010
Project Officer
Thomas E. Waddell
Technology Development and Applications Branch
Environmental Research Laboratory
Athens, Georgia 30613
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GEORGIA 30613
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
U.S. Environmental Protection Agency, Athens, Georgia, and approved for pub-
lication. Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
FOREWORD
As environmental controls become more costly to implement and the
penalties of judgment errors become more severe, environmental quality
management requires more efficient management tools based on greater know-
ledge of the environmental phenomena to be managed. As part of this
Laboratory's research on the occurrence, movement, transformation, impact,
and control of environmental contaminants, the Technology Development and
Applications Branch develops management and engineering tools to help
pollution control officials achieve water quality goals through watershed
management.
Pollutants in runoff and seepage from urban, agricultural, and forested
areas contribute significantly to water pollution problems in many areas of
the United States. Control strategies for reducing nonpoint source pollution
require a body of knowledge and set of analysis methods that reflect the
complex nature of the problem. This manual was developed as a guide for
water quality planners who must design cost-effective nonpoint source con-
trols for irrigated and nonirrigated cropland as part of areawide water
quality management strategies.
David W. Duttweiler
Director
Environmental Research Laboratory
Athens, Georgia
m
-------
ABSTRACT
This manual is designed to serve as a guide for the evaluation and
selection of agricultural nonpoint source controls. Such controls are
specified in water quality management plans developed in response to Section
208 of the 1972 Federal Water Pollution Control Act Amendments. Control
practices are described for both irrigated and nonirrigated cropland. The
evaluation methodology presented is applicable to areas ranging in size from
individual farms and sub-watersheds to large regions,.
The manual contains five sections including the introduction, and six
appendices. Section 2 presents the physical and chemical nature of nonpoint
source pollutants and the pathways by which these pollutants are trans-
ported from field to water body. Pollutants discussed include nutrients,
sediment, animal wastes, salts, and pesticides.
Controls involving tillage methods, cropping practices, and methods of
fertilizer and pesticide application are described in Section 3. Control
options are categorized according to the relative permanence of the
measures into three groups: structural, vegetative, and managerial.
Section 4 presents the actual evaluation methodology, which involves the
following seven steps: 1) Description of the watershed, 2) Identification
of the problem, 3) Determination of applicable control measures, 4) Choice
of the unit of analysis, 5) Establishment of the base condition, 6) Evalua-
tion of control measures, and 7) Development of an optimal control strat-
egy. Problem identification, Step 2, is actually beyond the scope of the
manual and is assumed to have been completed in earlier planning stages.
The end-product of these steps is a ranking of the cost-effectiveness of
specific practices controlling pollutant pathways in a region.
Two case studies described in Section 5 demonstrate the application of
the evaluation methodology to watersheds in Ohio and Washington States. A
single farm is evaluated as part of the Washington State case study.
The planning manual is supported by the following six appendices which
provide details of pollutant transport processes and methods of evaluation:
A. Estimating Nonpoint Source Sediment and Nutrient Loadings
from Non-Irrigated Croplands
B. Sediment and Nutrient Loss Estimates from Irrigated
Agriculture
C. Estimating the Effectiveness and Costs of Salinity Control
Measures
-------
D. Nonpoint Source Water Quality Problems Related to Animal
Agriculture
E. Water Quality Impact and Control Alternatives Associated with
the Use of Insecticides
F. Economic Perspective and Evaluation Methods for Agricultural
Nonpoint Source Water Quality Management
This report was submitted in fulfillment of Grant No. R804925010 by
Cornell University under the sponsorship of the U.S. Environmental
Protection Agency. This report covers the period February 1, 1979 to
February 1, 1980, and work was completed as of February 1, 1981.
-------
CONTENTS
Foreword i "> i
Abstract }y
Acknowledgements , vii
1. Introduction 1
2. The Problem 3
Agricultural Practices and Their Impact on Water Quality... 3
Guidelines for Water Quality Improvement 3
Point and Nonpoint Sources of Pollution 3
Natural (Base) Loads 4
Water Use Impairment 4
Physical Factors Affecting Degree of
Water Quality Impact 8
Stages of Pollutant Transport 8
Precipitation and Temperature 8
Topography 10
Soil Type 10
Cropping Practice 11
Irrigation Applications 11
Stream/Lake Characteristics 13
Agricultural Nonpoint Source Pollutants:
Factors Related to Their Control 13
Nitrogen 14
Phosphorus 15
Sediment 16
Animal Wastes 17
Salinity 18
3. Nonpoint Source Control Practices 20
Criteria for Grouping Nonpoint Source Controls 20
Candidate Measures for the Control of Nonpoint
Agricultural Sources 21
Non-Irrigated Agriculture 21
Management Controls 25
Vegetative Controls 28
Structural Controls , 29
Irrigated Agriculture 30
Management Controls 38
Vegetatative Controls 42
Structural Controls 42
Institutional Controls 46
VI
-------
4. Methods for the Evaluation and Selection of Agricultural
Nonpoint Source Controls 47
Step 1: Description of Watershed 47
Step 2: Problem Identification 47
Step 3: Determining Applicable Control Measures 50
Step 4: Choosing the Unit of Analysis 54
Step 5: Establishing the Base Condition 54
Step 6: Evaluating Candidate Control Measures 60
Step 7: Developing an Optimal Control Strategy 62
Evaluation of Pesticide Controls 68
Factors Influencing Successful Implementation of NPS
Control Practices 74
Summary 77
5. Examples of the Agricultural NPS Selection Process 78
Purpose of Case Studies 78
Case Study I: Honey Creek, Ohio 78
Description of Watershed 78
Problem Identification 79
Determining Applicable Control Measures 82
Choosing the Unit of Analysis 83
Establishing the Base Condition 83
Evaluating Control Measures 91
Developing an Optimal Control Strategy 93
Case Study II: Yakima River Basin, Washington 100
Description of Watershed 101
Problem Identification 101
Farm Model 102
Watershed Model Ill
References 122
Appendices
A. Estimating Nonpoint Source Sediment and Nutrient Loadings
from Non-Irrigated Croplands 124
B. Sediment and Nutrient Loss Estimates from Irrigated
Agriculture 326
C. Estimating the Effectiveness and Costs of Salinity Control
Measures 364
D. Nonpoint Source Water Quality Problems Related to Animal
Agriculture 430
E. Water Quality Impact and Control Alternatives Associated with
the Use of Insecticides 485
F. Economic Perspective and Evaluation Methods for Agricultural
Nonpoint Source Water Quality Management 655
Glossary 720
vn
-------
ACKNOWLEDGEMENTS
The compilation of material in this report would not have been
possible without the assistance of a number of individuals. Dr. Steven
Yaksich and Mr. John Adams of the U.S. Army Corps of Engineers, Buffalo
District, Dr. Terry Logan, Dr. Samuel Bone and Dr. Lynn Forster of Ohio
State University, and Mr. George Stem with the Soil Conservation
Service, Medina, Ohio all provided important data and suggestions for
case study analyses.
Ms. Karen McCombe, Cornell University, skillfully directed
editorial and format changes from which evolved tihe final report.
Mrs. Sandra Bates, Ms. Dorothy Clarke, Ms. Deena Dunn, Mrs. Sue
Fredenberg, Mrs. Terry Kinsman, Mrs. Marion Ogden and Mrs. Ruth Stanton
all contributed to the typing of the report.
Vlll
-------
SECTION I
INTRODUCTION
Excessive growth of aquatic plants in Lake Erie decreases game fish
populations, increases the level of treatment necessary for municipal drink-
ing water supplies, and discourages recreational use of the water. High
salt concentrations in the Colorado River make the water unfit for irriga-
tion purposes, drinking water, and other municipal/industrial uses. These
are both nationally publicized cases of serious, long-term water pollution
problems. The major source of pollution in these cases is not industrial
and municipal discharges, considered to be the cause of most serious water
pollution problems in the country. The source rather is runoff and seepage
from urban, agricultural, and forested areas. Pollution such as this, which
originates from an unconfined source not from a localized source such as a
pipe, is referred to as nonpoint source (NPS) pollution.
In 1972, the Federal Water Pollution Control Act Amendments (FWPCA)
were passed in response to a recognition of the seriousness of both point
and nonpoint water pollution problems. The national objective for water
quality improvement as set forth in section 101(a) of the 1972 Amendments,
is "to restore and maintain the chemical, physical and biological integrity
of the nation's waters." The act states that by 1985, pollutant discharges
into navigable waters from point and nonpoint sources shall be eliminated.
An interim goal of this act is to provide for the protection and propagation
of fish, shellfish, and wildlife, and to permit recreation in and on the
water, wherever feasible, by July 1, 1983.
Section 208 of the FWPCA amendments calls for areawide water quality
management plans to be developed and implemented by each state to insure
adequate control of both point source and nonpoint source pollution. These
plans must identify water quality problems and possible sources of
associated pollutants, and propose methods for pollution control. Although
the development and implementation of these plans is the primary
responsibility of each state, financial and technical assistance is
available from the federal government.
PURPOSE OF MANUAL
The purpose of this manual is to serve as a guide for the development
of agricultural nonpoint source controls called for in 208 water quality
management plans. Step by step procedures outlined in the manual provide a
basis for the cost-effectiveness ranking of specific practices controlling
pollutant pathways in watersheds. Criteria and computational methods for
evaluating different control strategies are presented. The manual is
1
-------
designed to be used in preliminary planning stages, not in the final stages
of site-specific design.
SCOPE OF MANUAL
The material presented in this manual is oriented towards water quality
planners involved with nonpoint source pollution control programs. It is
assumed that the planners are familiar with the problem and the geographic
area to be evaluated. Knowledge of typical farm enterprises in the region
and the engineering feasibility of control practices is assumed. The plan-
ning team may have to consult with various federal, state, or local groups
to gain this familiarity. The Soil Conservation Service (SCS/USDA) and
State University Departments of Agronomy, Agricultural Engineering, and
Agricultural Economics can provide valuable assistance.
The methodology presented in this manual is appropriate for the design
of nonpoint source controls for all irrigated and nonirrigated cropland. It
can be used to evaluate areas ranging in size from individual farm fields to
large regions. With such broad applicability, specific soil, crop, pre-
cipitation and other physical as well as economic data are not supplied. The
case studies presented in Section 5 demonstrate how the methodology is
applied in specific situations. The accompanying technical reports provide
details of computational methods and data collection procedures.
This manual focuses on three key pollutant categories: crop nutrients,
sediment, and salts. Methods of estimating losses of these pollutants are
presented. Livestock waste and pesticides are covered as separate pollutant
categories. Evaluation methods for estimating losses of these pollutants
are less well developed, thus these pollutants are treated in less detail.
USE OF MANUAL
Because nonpoint source pollution is both a relatively recent concern
and a complex phenomenon, many unknowns remain. The extent to which agri-
cultural sources contribute to the total pollutant load, the extent to which
various control practices decrease this load, and the effects of reduced
pollutant loads on water quality are generally not easily determined. The
manual makes certain assumptions concerning how this information is to be
derived. The manual assumes that:
1) Water quality problems have been defined and. associated pollutants
have been identified in previous stages of planning,
2) The relative contribution of agricultural sources to the total
pollutant load has been estimated,
3) The pollutant load delivered to the waterbody will be reduced by
reducing edge-of-field losses,
4) Physical and economic mathematical models used to estimate cost-
effectiveness of practices are sufficiently accurate to use for
the purpose of comparing the relative cost-effectiveness of dif-
ferent control strategies, and
5) Some field data collection may be required. The manual lists data
required for the various steps of the evaluation, and indicates
where field studies may be necessary.
-------
SECTION 2'
THE PROBLEM''
AGRICULTURAL PRACTICES AND THEIR IMPACT ON WATER QUALITY
Guidelines for Water Quality Improvement
The Environmental Protection Agency (EPA) Guidelines for State and
Areawide Water Quality Management Program Development (EPA, 1976) suggest
that "Best Management Practices" be identified and implemented for categor-
ies of pollutants. A Best Management Practice (BMP) is defined as "a prac-
tice or combination of practices that is determined by a state (or desig-
nated areawide planning agency) after problem assessment, examination of
alternative practices, and appropriate public participation to be the most
effective, practicable means (including technological, economic, and insti-
tutional considerations) of preventing or reducing the amount of pollution
generated by nonpoint sources to a level compatible with water quality
goals". EPA advises that wherever possible, practices should concentrate on
controlling each pollutant at its source. Since collection and treatment of
polluted water is generally complex and expensive, treatment is to be
resorted to only in situations where preventive measures will not result in
the attainment of desired water quality standards.
Cost sharing for the implementation of BMPs was authorized by the Rural
Clean Water Act of 1977. BMPs to control nonpoint source pollutants are
installed and maintained under 5-10 year contracts with cooperating land
owners.
Point and Nonpoint Sources of Pollution
Whereas point source pollutants have a localized identifiable source,
typically a pipe, nonpoint source pollutants originate from an unconfined
source, typically a relatively large area of land. Sources include land
used for agriculture, silviculture, mining, and construction. Urban land
over which stormwater flows is another source. Transport of pollutants from
nonpoint sources to water bodies is by overland flow, percolation, and
erosion and sedimentation. Pollutants which may originate from agricultural
land include nitrogen, phosphorus, sediment, manurial organic matter, salts,
and pesticides.
The unique characteristics of nonpoint source pollution pose special
problems which require a body of knowledge and set of methods very different
from point source pollution problems. Whereas identifying the origin and
measuring the discharge associated with a point source involves well-estab-
-------
1ished techniques which are accurate and easily applied, identifying and
assessing the magnitude of nonpoint source contributions requires new and
very different techniques.
The complexity of control strategies required for nonpoint source pol-
lution reflects the complex nature of the problem. Nonpoint pollutants are
not transported or delivered to water bodies in a confined manner, there-
fore there is no efficient and economical means of monitoring. The use of
effluent limitations is thus generally not applicable for nonpoint source
control. Control strategies, on the other hand, must be developed from a
consideration of the many variables controlling pollutant transport. An
understanding of the dynamics of chemical cycling in soil and water is
required to effectively identify nonpoint sources, assess impacts, and
develop appropriate control strategies.
Natural (Base) Loads
Many pollutants, including nitrogen, phosphorus, organic matter, patho-
gens, salts, and sediment are present to varying degrees in all aquatic
systems. It is important to estimate the magnitude of natural pollutant
loads before implementing control measures. Where the base load is rela-
tively large, the improved management of adjacent agricultural lands may
not significantly improve water quality.
Water Use Impairment
Although nitrogen, phosphorus, organic matter, salts and sediment are
natural components of all aquatic systems, in excess they can offset criti-
cal environmental balances. Table 1 summarizes how nutrients, sediment,
animal wastes, salts and pesticides can affect the quality of receiving
waters.
Nutrients--
Nutrients, such as nitrogen and phosphorus, are generally present at
relatively low concentrations in the aquatic environment (i.e., below 0.3
and 0.05 ppm for nitrogen and phosphorous, respectively). When man augments
the quantities of nutrients in a stream or lake, aquatic productivity can be
increased dramatically. This process, referred to as cultural eutrophica-
tion, may reduce the value of the water to humans. Decaying organic matter
may produce unpleasant odors and deplete the oxygen supply for aquatic
animals. Excess plant growth may interfere with recreational activities
such as swimming and boating. Decreased oxygen levels, especially in the
cold bottom waters where decaying organic matter tends to accumulate,
reduces the quality of game fish habitats. The lake or stream waters can
become turbid and noticeably colored. When used as a water supply by
municipalities and industries, the level of pre-treatment often must be
increased.
Cultural eutrophication is not the only water quality problem caused by
nutrient enrichment. Dissolved ammonia at concentrations of more than 0.02
ppm of NH3 may be toxic to fish. Nitrates in drinking water are potentially
-------
i
_j
d
a.
to
|
CXL
>-
0
to
s
£
1
o.
*—
UJ
S
at
Ul
i
UJ
CO
h-
«
i
« u
p
fr*.
flo
a.
^jf
Q_ *""
S
V) 3
O O
U- O.
o
c >>
£3
. ftf 14.
-0 >£ 0
C **- IO 4-> VI
2-5 °" * cn
.c "Jo o •* -a •"
w t. • u ai •»-»
at *j o -M *
U C -O l/> 10 3
.c 4i c vi (- cr
c
— u
U) C -C ID 4- O
oioi-u->ua>o4-> u a.
O»-i- O *T C C •-- * O
itJ-MM-ACnmC-r- JO O> Ul
r- 1. IB •— C O O C U C -C
i— 3 •— +J 3 Cn-*- *O J- .C i1'-Q- T3 *J"DinO
CCOECCi- CXOJ<~ Ci~S~
OCOIQ.-P003:OE O Cr— > C -M
uo^ai Q.OIL. co«+-3 a'^-TS-M
ID Ol/lt->»0)P~Vl^'*- -r--f-OV> t-OC
•r- •— - CD •»- 4-> f— 4J ID "*-. •*-" Q C i- —
O Q. C 'OD.l-'OUlOai *— E O C i. •- 1- Q.T3
EO.-'-OJO»34-»'*--r--f-UC 3 t. -r-
5eMOlS«COI 0.10"- 0)*t-0ifl is) C E 0) •<-
OjC(7>CnOIL-^L.L.+J4J L.V>30) >, r- TO (J >,
j: • vi c i i- i- ot x: oiOv>3 c-ac roeais
f^Sl^sl-SlsIl aSli Sssll
c
O
»— C 3
ai g («-
»— 3 *5
Stss ti.
x o ai o •— o
O *— 4-» -i- C i—
O (0 •*-" 3 O
•o u u n vi u
0) -4-* S- Ol V.
O VI -r- 4-> (S1V1 U4->O
k. V> V) UJ •*-
Ol£ Ol C J <4- Q j; ID o 1/1 i— m^-oioi -_ J C -S ^
(JOV •<- -r- ^ H- (— 3O l/)OX Ol (O Q.
•iJCTOIW *O>+J C *O4-' Z iciD-*-1! — CU
f+jai cnl-oi-'-'ew if-oi >-c:oia- v>
C**-r— ID 4-> VI O)O- OJ- Q£ -t- -^- r— H- O fc- 3 OJJ^1*-
«l-(O C fc. in 1-WO
0 i— 01 C -^UOli- vivlOl ID «0<+--0*-> i*-^
ll C -r- UulT3(O-t->OlV)a.Vl^fc. OlL-iQCOlO) £ O
C(J-'-+J "O-MCjr— W>Otg O'r-VlOl (/^l/)UO> •»- wl +J Cu r-
'•|3j=a> vioi v)t^ai->-o.oi u ro 01 i- 3 «— *o 01 3 j§ •£ •£ o vi
Ol iS)Q) U J2 t-4->Q13Z UJL- H3 0> •*-> S- iDCt-Ol
to cr-c ot '^'a'aJr— CC3CCOJI— Oi>4- 5 oicoioioco ra CT-— o r-
...
E,
o
o *->
H- 0-
ZZ Q. ^ 0.
3E i t Q- I "a O +J
I 01 rO I O) "O «-
|Q L. O ID ^ Q.-I- OOQUUO
Oi-EOO DO CDt-JZU-lj-
K- Z S K- (/) LO CO
en
c
"O XJ
c tr
to uj vi E •— 6
3 J— 3 Ol Cl-C3''-C E
LOOIO eCO)Ot-O> O
h- cnj= I- 3 oi£ c 01 en >- u
zoo. z oo.ai4->o »—
1— 1 4J O Z < •*-* O >v3L +J Z r~
^—ZQ- Q H^ZQ-OQ- _Ji-O
=) UJ Z <
JT
+J
S 0§
c o>
O 01 4->
iD 3 E
•r- C O
C .C 01 >
CTi U *J O)
r- O O C
£
£
(J I .
«*-
(LI
fp n
Oi -
JD
2
+-1 V.
0 4^
O fT3
U**- ^
01 •*-> ^ d
1- C VI
J Ol r- U
to O TJ
>! 01 -M 3
8*0. r~ ^
1- O 3 O C
(
- VI VI 0 "? S
-2 QJ "^ "a. c ^ ^
o'- '^^S'o"2™-S£
i- 'OCi-.col-'-^
O O'-Oa.'-t—UQ.
ai QJ v
a G u --
1 — fl> 3 •-
UJ
Q-
-------
toxic to humans, especially young infants, since their digestive systems are
less acid than those of adults and contain bacteria capable of reducing
nitrate-nitrogen to nitrite-nitrogen. Nitrite accumulation in the digestive
tract can, in turn, cause brain damage or even death, by reducing the
oxygen-carrying capacity of the blood (methemoglobinemia). Consequently,
the Environmental Protection Agency (EPA) has set a limit of 45 ppm nitrate
for drinking water. Ruminants can also be affected by nitrates via the same
chemical processes. Tolerance levels for ruminants have been set somewhat
higher than for humans.
Sediment--
Sediment affects the use of water in a variety of ways. Total sus-
pended solids can reduce the amount of sunlight available to aquatic plants,
cover fish spawning areas and food supplies, and clog the gills of adult
fish. This reduces fish, shellfish and plant populations and decreases the
overall productivity of a lake or stream. Recreation is limited because of
the decrease in fish and shellfish populations and because of the water's
unappealing, turbid appearance.
Sediment also fills road drainage ditches, culverts, and stream
channels and shortens the economic life of reservoirs and farm ponds. It
can plug water filters, erode power turbines and sprinkler nozzles, and
damage pumping equipment. Maintenance costs are increased and additional
treatment may be necessary before the water can be used for drinking or
industrial purposes.
Much of the sediment transported to waterways originates from agricul-
tural land. Valuable topsoil is thus removed and an increased application
of nutrients is required to maintain soil productivity.
Chemicals such as pesticides, solid phase phosphorus, ammonium and
organic nitrogen are also transported with sediment in an adsorbed state.
Changes in the aquatic environment can cause the release of these chemicals
from sediment. Thus, although adsorbed phosphorus transported by the sedi-
ment may not be immediately available for plant growth it may serve as a
long-term and continual contributor to lake eutrophication.
The reverse may also occur; chemicals already present in the water can
be scavenged by suspended solids. A pesticide adsorbed to sediment will not
necessarily become less toxic, but much of it will settle out of solution at
least temporarily.
Animal Wastes--
Animal wastes can contribute nutrients, oxygen-demanding materials, and
pathogens to receiving streams. The composition of most manures is such
that if enough is applied to meet the nitrogen needs of a crop, phosphorus
will be applied in excess of crop needs. Surface runoff from fields to
which manure has been applied thus may carry substantial quantities of
phosphorus. Leaching of soluble or organic forms of phosphorus from manure
application sites is generally negligible because of the soil's capacity to
adsorb phosphorus.
6
-------
The oxygen demand exerted by carbonaceous materials individually or in
combination with nitrogen can deplete dissolved oxygen supplies in water,
and may result in anaerobic conditions. When the decomposition process
becomes anaerobic, methane, amines and sulfide are produced in addition to
the carbon dioxide, sulfate, ammonia and nitrates that result from aerobic
decomposition. The water acquires an unpleasant odor, taste and appearance
and becomes unfit for drinking and recreational purposes. Treatment is
often required for industrial use.
Salinity--
Water used for irrigation accumulates salts. In some areas of the
country, water and land quality has been seriously degraded due to salin-
ity. The total salt burden of western streams may be as much as 40 percent
man-caused (Law and Bernard, 1970).
Excess soil salinity can delay or prevent crop germination and substan-
tially reduce the rate of plant growth. High osmotic pressures that develop
between the soil solution and the plant root impair the plant's ability to
absorb water. Other possible adverse effects of salinity include nutri-
tional imbalances, toxicities caused by specific ions such as boron which
can be toxic in small quantities, and poor soil aeration and water transmis-
sion if excess sodium accumulates and disperses the soil. The addition of
saline irrigation water to surface water bodies limits potential downstream
usage for drinking, irrigation or industrial purposes. High salt concen-
trations in streams and lakes can harm fresh-water aquatic plants just as
excess soil salinity damages crops, although salt levels rarely reach such
levels except in land-locked bodies such as the Salton Sea (50,000 mg/1) or
the Great Salt Lake (270,000 mg/1).
Pesticides--
Pesticides include insecticides, herbicides, mitacides, nematocides,
rodenticides, fungicides, plant growth regulators, and desiccants, all of
which are used extensively in agriculture and silviculture. The utilization
of pesticides in agriculture has greatly increased since they were first
used in the 1940s. According to USDA, in 1976, 1015 million pounds of pes-
ticides were used in this country. Of this quantity, about 65% were used on
farms (von Rumker, et al., 1975). Approximately 56% of all nonpasture
cropland was treated with herbicides and 18% was treated with insecticides.
Although the benefits of pesticide use are substantial, there are also
environmental risks related to its use. Some types of pesticides or their
metabolites are resistent to degradation. Degradation products may persist
and accumulate in aquatic ecosystems. The entire food web including man can
be affected.
Sublethal effects include those behavioral and structural changes in an
organism which jeopardize its survival. For example, certain pesticides
have been found to inhibit bone development in young fish or affect repro-
duction by inducing abortion.
-------
Herbicides typically degrade quickly, however can destroy non-target
vegetation in the aquatic environment that serves as a food source for
higher organisms. Indirect effects of herbicide use thus include the
disruption of food chains and the addition of organic material to waters
promoting eutrophication.
Biomagnification is a phenomenon which occurs if a pesticide is taken
in by an organism but not excreted. An organism will accumulate a higher
pesticide concentration than is present in the organisms upon which it
feeds.
Physical Factors Affecting the Degree of Water Quality Impact
Stages of Pollutant Transport--
The process by which a substance is delivered to a stream can be des-
cribed in terms of three stages: availability at the field site, detachment,
and transport (Figure 1). Only when a pollutant is available in some form
at a site, becomes detached, and is transported to a receiving body does it
constitute a potential pollution hazard.
Factors which determine the nature of the pollutant delivery process in
a given situation include precipitation, temperature, topography, soil
type (including antecedent moisture conditions), cropping practices and
irrigation applications. Physical characteristics of the receiving water
body affect the actual impact of the pollutant after it is delivered.
Precipitation and Temperature--
Temperature, rainfall pattern, and snow melt are important climatic
variables that affect infiltration, surface runoff, and soil erosion. At
lower temperatures, the rate of nutrient uptake by plants decreases. More
nutrients in the soil may thus be available for detachment and transport,
however less nutrients will be taken up by plants in the receiving water
bodies. The effects of any increase in nutrient transport are further off-
set during cold weather by the capacity of cold water to hold more oxygen
than warm water.
High intensity storms generally increase the runoff of water and the
dislodgement and transport of pollutants. Over shorter periods of time (for
the same total rainfall) the water will have less opportunity for infiltra-
tion. The high kinetic energy of raindrops causes larger soil particles to
be dislodged from the surface and transported with runoff waters and smaller
particles to become cemented, sealing the soil surface. Raindrop impact of
high intensity storms also tends to seal the soil surface.
Melting snow can contribute significantly to the runoff-erosion pro-
cess. An increase in runoff, which often results from rapid snow melt,
provides greater energy to dislodge and transport soil and associated
adsorbed substances and increases the volume of water for transporting
soluble materials.
-------
fO
O)
S-
4->
to
o
O
O
Q.
O)
O)
O)
Id
cr
i—
CO
O
I-
Q
_l
UJ
>-
O
a
cr
o
a.
CO
z
<
cr
i-
LJ
2
I
UJ
Q
CD
4
-------
Topography--
The capacity of water to dislodge and transport material increases with
its velocity. As water moves downslope its velocity increases. The steeper
the slope, the faster the increase, and the longer the slope, the greater
the water volume and velocity at the bottom. With higher velocities, ero-
sion and overland flow are promoted and infiltration inhibited.
Shape of the slope is also important, with concave hills generally
delivering fewer pollutants to a stream or lake than convex hills. Trans-
port capacity decreases at the bottom of a concave slope, allowing some of
the suspended pollutants to settle. The reverse is true of convex hills.
Soil Type--
The rate at which water infiltrates soil affects the ratio of surface
to subsurface flow. With an increase in the infiltration rate of a soil,
the pollutant load associated with runoff should decrease. Both the solu-
bility and adsorptive nature of a pollutant will determine whether it be-
comes adsorbed to soil particles or percolates through the soil.
A soil's infiltration rate and its ability to adsorb pollutants depends
on soil characteristics. Prior moisture content markedly affects the amount
of water which can infiltrate into a soil. The amount of organic matter and
clay particles largely determines the adsorption capacity of the soil.
Sandy soils generally have high infiltration rates, because the large soil
particles result in relatively large pores, through which water can perco-
late readily. Because total surface area and the negative charge of these
soils are less, their adsorption capacity is generally much less than clay
soils. Soils that are both well drained and contain a sufficient amount of
clay and organic matter will absorb the most pollutants. Subsoil character-
istics may either retard or enhance drainage, thus influencing the parti-
tioning between surface and subsurface flow.
Soil fertility can influence pollutant flux. An increase in fertility
may increase evapotranspiration due to increased plant growth. This will
decrease both runoff and subsurface flow. The increased crop canopy will
provide for greater protection of the soil surface from raindrop impact, and
the larger, deeper root system will increase soil stability during runoff
events. The eroded soil, however, will have a greater content of nutrients.
Soil type affects the polluting potential of an area by limiting the
fundamental choice of cropping practice and/or irrigation system options.
For example, poorly drained soils require certain cultivation practices and
often allow only limited cropping choices. Excessively drained soils
require shorter, more frequent irrigations, and frequent fertilization, and
often cannot be efficiently irrigated with conventional surface irrigation
methods.
10
-------
Cropping Practice--
Crop cover protects the land from rainfall impact, thereby decreasing
soil erosion and increasing infiltration. In general, the potential for
soil erosion and runoff will be smaller the denser the crop cover, the
longer the crop is on the land, and the greater the quantity of residue left
after harvesting.
Root growth increases soil stability. In addition, crops with deep
roots generally require less frequent irrigations. Both enhanced soil
stability and reductions in irrigation frequency help to minimize pollutant
losses from croplands.
Certain cropping practices can reduce the need for pesticide applica-
tions. For example, using more resistant crop varieties may allow reduced
pesticide application rates, and mechanical weed control techniques may
decrease the need for herbicide use.
Irrigation Applications--
In arid regions, which dominate much of the western United States,
agricultural land is commonly irrigated. Irrigation not only increases
productivity (by 2- to 3-fold), but also provides flexibility by allowing a
farmer to grow a variety of valuable crops such as corn, cotton or sugar
beets. However, as irrigation water moves across and through the furrow-
irrigated fields, salt concentrations are increased through evapotranspira-
tion, and other pollutants are leached from the soil.
The degree of impact that leachate and return flows have on the
quality of receiving water is determined to a large extent by the method of
irrigation. Irrigation volume and efficiency will vary markedly among
furrow, basin, border, sprinkler and trickle irrigation methods. Given a
specific method, different system design and operating practices result in
widely different efficiencies. For furrow-irrigated fields, for example,
the magnitude of runoff and percolation depends on stream size (quantity of
water delivered per unit time to the head of each irrigation furrow), furrow
slope (steepness of the irrigation channel) and length, frequency of irriga-
tion, set time (length of time each furrow is irrigated during a given irri-
gation), uniformity of water application, and intake rate (Figure 2).
An increase in either stream size or furrow slope generally increases
the detachment and transport capacity of the irrigation flow. Flow
decreases as the furrow is traversed, however, so midfield deposition of
sediment eroded from the head of the field is common. Usually irrigators
reduce stream size as the slope is increased, but the reduction is rarely
enough to compensate completely for the slope increase. Set time is often
determined by tradition or convenience to the operator, though longer set
times coupled with less frequent irrigations generally lead to decreased
losses of sediment and adsorbed pollutants. Increased set times, however,
may increase percolation.
11
-------
Frequency of Irrigation
Set Time
Uniformity of
Furrow Applications
Leaching
Return
Flow
Figure 2. Variables influencing pollutant losses
from irrigated fields.
Furrow irrigation is characterized by considerably less uniformity of
water application than sprinkler irrigation. Because there is a nonuniform
depth of application, the soil below the furrows receives more water than do
other areas of the field. Long furrows (larger lengths of run) further
decrease uniformity, because the portion of the field near the lower end of
the furrow receives a smaller stream flow for less time than do those areas
near the upper end of the field. Consequently, less water infiltrates the
lower portion of the field. Nonuniformity is disadvantageous because it
results in over-application to some portions of the field and/or under-
application to other portions. Over-irrigation causes excessive percola-
tion, whereas under-irrigation causes plant stress and decreased crop
yields.
12
-------
Stream/Lake Character!'stics--
The quantity of pollutants that may be adequately assimilated by a
water body is determined by many factors which cannot be easily controlled.
Physical characteristics of the receiving waters, such as size, rate of flow
(turnover rate), base load, temperature, and pH all affect the degree to
which a substance actually impairs water quality (Figure 3). The larger the
volume of the receiving water and the shorter the turnover rate, the less
sensitive the water body will be to an influx of pollutants. Volume and
turnover rate are especially important in irrigated regions. If the base
stream flow is large, the pollutant concentration which results after
irrigation return flows have been mixed with the stream may not be large
enough to adversely affect usage.
Figure 3. Factors influencing the assimilative capacity of waterbodies.
Base Load
Load from
Non-Point Sources
Point Source Load
Water Quality
Impact
Temperature
Size
Turnover Rate
Characteristics of Lake
Bottom Sediments
AGRICULTURAL NONPOINT SOURCE POLLUTANTS: FACTORS RELATED TO THEIR CONTROL
A nonpoint source control can be evaluated in terms of both the stage
of the pollutant delivery process during which it is operative and the
pollutant pathways it affects. A pollutant may undergo many physical and
biological transformations between the time it becomes available in the
13
-------
field and the time it is delivered to a stream, lake, or groundwater reser-
voir. Substances initially strongly adsorbed to soil particles, for exam-
ple, may later be detached and become dissolved in runoff waters. The
transformations that a pollutant undergoes determines which pathway(s) it
will take between cropland and receiving water. A pollutant can be moved
suspended or dissolved in water, or attached to soil particles in overland
or subsurface flows. Figure 4 illustrates the pathways pollutants may
follow as they move to a water body.
Figure 4. Pollutant pathways.
Drift
Percolation1 Volatilization
Overland
SubsurfaceN^Flow
Flow
Base Flow
Evapotronspiration
s
Nitrogen
Although nitrogen (N) is naturally present in soils, it must be added
to most croplands in order to increase their productivity. Manure spread-
ing, application of commercial fertilizers, and incorporation of crop resi-
dues all contribute nitrogen to the soil. Not all of the nitrogen present
on or in the soil at a given time is available for plant uptake. Organic N
(primarily in particulate form) normally constitutes the bulk of soil nitro-
gen. It is only slowly converted to the more readily available inorganic
forms of nitrogen (inorganic ammonium and nitrate).
14
-------
Because of its mobility in soils, nitrate is the form of nitrogen most
commonly used by plants. Plants may also use ammonium nitrogen as an addi-
tional source.
Nitrate-nitrogen is highly soluble and will readily percolate below the
crop root zone. It may also be transported with surface runoff, but not
generally in high concentrations. Organic-nitrogen and ammonium, on the
other hand, become adsorbed to soil and are lost primarily with eroding
sediment. Organic nitrogen generally converts to inorganic nitrogen with
time, thus all transported nitrogen is a potential contributor to eutrophi-
cation.
Three microbial transformation processes relate to the control of
nitrogen. The first two, ammonification and nitrification, are part of the
mineralization process which converts nitrogen from its organic to its
inorganic form, making it available for crop uptake. The last reaction,
denitrification, causes nitrogen to be lost to the atmosphere. Denitrifica-
tion may reduce the quantity of nitrogen lost via percolation and runoff,
however, it also means that some fertilizer is wasted.
Nitrogen may be lost in the gaseous form by processes which do not
require the presence of microorganisms and are strictly chemical. Such
losses of nitrogen as ammonia can occur especially with heavy surface
applications of urea or ammonium fertilizers.
Practices which cause anaerobic soil conditions promote denitrifica-
tion, especially when accompanied by a source of soluble carbon to serve as
an energy source for the denitrifying organisms. Such conditions are
especially prevalent at waste application sites and beneath animal feed lots
and corrals. Practices which increase either the oxygen content of the soil
or soil drainage (to remove the products of mineralization) increase miner-
alization, thereby allowing more nitrogen to become available for plant
growth or for leaching. Also, if fertilizers or manure are not incorporated
into the soil, the potential for nitrogen loss via volatilization to the
atmosphere and via surface runoff will be greater. Finally, the timing of
fertilizer and manure applications to meet crop needs is critical. The
closer an application is made to the period of maximum crop growth, the
smaller the potential for losses. For example, the use of split applica-
tions, at time of planting and just before the peak growth period, is
usually recommended.
Phosphorus
The total phosphorus (P) content of most soils is low, between .01 and
.20 percent by weight (Brady, 1974). Since much of this phosphorus is un-
available for plant growth, manure and fertilizers are used to increase the
level of available phosphorus in the soil. Some of this applied phosphorus
can reach nearby water bodies with runoff and erosion. In addition, phos-
phorus that may be unavailable in the soil system may still erode with soil
particles and become available when the bottom sediment of a stream or lake
becomes anaerobic. It may then cause water quality problems. Estimating
the potential impact of phosphorus loss on water quality is difficult at
15
-------
present, because the relationships between various forms of phosphorus in
the soil and sediments, water, and biota are only poorly understood. The
dynamics of phosphorus adsorption by bottom sediment are particularly ill-
defined.
Phosphorus can be classified as either organic or inorganic. Available
inorganic phosphorus, which serves as the intermediary between the bulk of
the total P, the organic P and the soluble fraction immediately available to
the plant is of most concern. Inorganic phosphorus can be either dissolved
in surface or subsurface waters or associated with sediments (particulate).
Although much of the particulate fraction acts as if it were permanently
fixed by the soil, some of the particulate inorganic phosphorus (the labile
fraction) serves as a source of the dissolved form. The labile pool is com-
monly several hundred fold larger than the soluble fraction at any point in
time. The fracton of the total particulate inorganic phosphorus which is
labile is generally not known. The equilibrium between labile particulate
and the dissolved inorganic phosphorus depends in part on the chemical and
biological characteristics of the limnological system. The amount of dis-
solved phosphorus changes during transport from cropland to streams and
lakes as well.
Losses of particulate inorganic phosphorus from croplands occur via
erosion since this form is associated with sediment. Rainfall events tend
to cause a higher particulate phosphorus loading of streams than do snowmelt
events because the former tend to be more erosive.
Although dissolved inorganic phosphorus is found in all surface and
subsurface flows, a decrease in runoff may not necessarily decrease the rate
of dissolved inorganic phosphorus loading to a particular stream or lake.
Labile particulate inorganic phosphorus already present in the lake or
stream sediments may continue to be converted to soluble form. In other
words, a decrease in runoff may decrease total phosphorus loss but may have
little short-term effect on the levels of available phosphorus in receiving
waters. If the amount of pre-existing labile phosphorus in bottom sediments
is small, however, then controlling runoff may reduce dissolved phosphorus
concentrations and substantially improve water quality.
Sediment
Eroded soil may either be redeposited on the same field or transported
from the field in runoff waters. Sediment is that soil which leaves a field
and is subsequently deposited in streams, lakes, roadside ditches or other
off-field areas. In nonirrigated regions, the sediment delivery process
starts by the detachment of soil particles either as a result of raindrop
impact or overland flow. Erosion can be classified either as interrill or
rill erosion. Interrill erosion is that erosion caused mainly by raindrop
impact and subsequent shallow flow toward rills. It is thus independent of
slope position and occurs over relatively small areas. When the runoff from
interrill areas becomes sufficiently concentrated, small but well-defined
channels or rills are formed. Rill erosion can contribute substantially to
suspended sediment loads since channelization of water increases flow velo-
city, which in turn increases the ability of the runoff to dislodge and
16
-------
transport soil particles. Gully erosion is an advanced stage of rill
erosion where field operations are often impaired.
In irrigated regions, soil particles are detached both by raindrop or
sprinkler drop impact, and by the surface application of irrigation water.
Significant amounts of soil are eroded by the flow in irrigation furrows
since the volume and velocity of such flows are relatively large. These
flows not only contribute to the detachment process, but also provide the
means for transport of detached particles. Because irrigation water per-
colates along the length of the irrigation furrow however, flow volume and
velocity decrease and mid-field deposition of sediments eroded from the head
of the irrigated field is common. Much of the sediment eroded from
furrow-irrigated fields actually originates in the bottom 1/2 to 1/3 of the
field, particularly if water velocity increases toward the end of the field
as the streams approach a relatively deep return flow collection ditch.
Sediment from cropland usually contains a higher percentage of finer
and lighter particles than the soil from which it originates. Although
large particles are more readily detached from the soil surface because they
are less cohesive, they will also settle out of suspension more readily.
Clay particles and organic residues on the other hand, once detached, will
stay suspended for longer periods of time and will be transported more
readily in slower flowing water. They are thus apt to be transported for
greater distances, and a larger portion of them will leave the field. This
selective erosion process can increase overall pollutant delivery, because
small particles have a much greater adsorption capacity for other pollutants
than do larger particles.
The quantity and type of adsorbed pollutants in sediment is largely
determined by the nature of the soil from which it originates, and the type
of erosion. Rill and interrill erosion moves soil particles from the sur-
face (plow) layer of the soil. Gully and streambank erosion can move par-
ticles that were part of the lower soil strata as well. Topsoil is usually
richer in nutrients and other chemicals than the subsoil because of normal
nutrient cycling in the soil profile and past fertilizer and pesticide ap-
plications. Topsoil is also more likely to have a greater percentage of
organic matter. Sediment originating from surface soils thus will have a
higher total chemical enrichment than sediment from gullies or stream banks.
Sediment delivery can be reduced by either controlling detachment or
transport. Since clay and organic matter are easily transported, it may be
easier to control these potential pollutants through practices that control
detachment. Sands or coarse silt particles are very easily detached, thus
Animal Wastes
Manure is a source of organic matter, nutrients and pathogens. Manure
applied on the soil surface will be more easily lost in runoff than manure
incorporated into the soil. As discussed in the sediment section, organic
matter is not easily detached from the soil surface because of its cohesive
properties, but it is easily transported because of its low density. There-
fore, practices which control detachment will probably be more effective in
reducing the loss of organic matter than practices which control transport.
17
-------
Losses of manurial organic matter can also be reduced by controlling
the rate, location, and timing of applications.
Although animal diseases may be transmitted to humans through contact
with animal feces, applied manure Is rarely a public health problem. The
pathogens present in manure are filtered by soil particles and rarely infil-
trate farther than a few centimeters into the soil profile. Consequently,
fecal ground water contamination is usually not a problem following animal
waste application to soil. Although erosion and runoff can detach manure
particles and transport them to streams and lakes, the number of pathogenic
organisms reaching surface waters is generally limited and their survival
rate low. Low soil moisture levels, low pH values, high temperatures, and
direct solar radiation can all cause pathogenic populations to decrease
rapidly with time. Manure storage facilities may represent an exception,
since the manure-slurry environment enables the survival of many organisms.
Composting of the wastes prior to application is normally quite effective in
decreasing the numbers of active pathogens in the material.
Two aspects of animal waste disposal that have received special atten-
tion are problems related to runoff from barnyards, and winter spreading of
manure. Both are suspected to be relatively concentrated sources of nutri-
ent loading to surface waters. Barnyards are areas of intensive use by
livestock characterized by the absence of vegetative cover and the presence
of appreciable accumulations of manure. Direct runoff through the barnyard
from upland drainage areas and direct precipitation on the barnyard will
detach and transport animal wastes from the disposal area. The wastes con-
tain nutrients, and in addition, are characterized by high biological oxygen
demand (BOD) and nitrogenous oxygen demand (NOD), thus if delivered to a
water body, will impact water quality.
Spreading manure on frozen ground or snow cover can lead to relatively
high concentrations of nutrients during subsequent runoff events. Such high
losses have been generally associated with critical periods such as spread-
ing on saturated ground after a rainfall, or on actively melting snow.
Salinity
The accumulation of salts in agricultural soils is an undesirable con-
sequence of supplying the land with irrigation water needed to sustain or
enhance agricultural production. Since irrigation water is generally de-
rived from ground water supplies or river sources, it has a natural (base)
load of dissolved salts. As Irrigation water is consumed by the plants or
lost to the atmosphere via evaporation, salts (and other pollutants) remain
in the soil. High salt concentrations are created within the soil profile,
especially in the root zone where most water removal occurs. Salt crusts
may form on the soil surface between irrigation furrows or where the land is
left fallow, and accumulations of soluble and exchangeable sodium lead to
soil dispersion and structure breakdown. Increases in soil and water sal-
inity can be quite substantial, since 70 to 80 percent of the applied water
is lost through evapotranspiration. Fortunately, a tripling or quadrupling
of the salt concentration of most irrigation waters still leaves soil solu-
tions which are not lethal to the growth of normal crop plants.
18
-------
In order to maintain productivity in irrigated agriculture, accumulated
soil salts must be moved periodically below the root zone to prevent the
impairment of plant growth. Therefore, the quantity of water diverted for
crop use must exceed actual plant water requirements to allow for the leach-
ing of these soluble salts. Although this procedure promotes improved plant
growth, it also increases the environmental impact of irrigation by increas-
ing the transport of salts and mobile nutrients (e.g. nitrate) to receiving
waters. As excess water percolates downward, its salt load is increased by
the leaching of natural salts arising from soil mineral weathering, atmos-
pheric deposition, or former marine or lacustrine (lake-deposited) sub-
strata. If the quantity of leaching water is not excessive, some salts may
react with other ions in the soil and precipitate, lowering once more the
salt load of the percolating waters.
Seepage losses from unlined delivery canals and laterals are high in
many irrigated areas. The combination of these seepage losses and deep
percolation losses can cause groundwater levels to rise near the soil sur-
face (water logging). Water and salts are supplied to the root zone by up-
ward movement of groundwater due to capillarity. The water moves to the
soil surface and evaporates, leaving its salts behind. This process can
result in extensive soil salination if allowed to operate for any apprec-
iable length of time.
Irrigation return flows provide the vehicle for conveying accumulated
soil salts to receiving streams or groundwater reservoirs. If the salt con-
centration of the return flow is small in comparison to total river flow or
groundwater capacity, water quality may not be degraded to the extent that
use is impaired. However, the process of withdrawing water for irrigation
and the return of saline water is frequently repeated many times along the
course of a river. Eventually, the salt concentration of the water can
become high enough to impair water use for irrigation or other purposes.
Salinity control is complicated because salts are present both in
receiving waters and in contributing soils, and also because control prac-
tices may interfere with crop production. The quantity of irrigation water
should not be decreased without increasing the efficiency of application, if
the level of production is to be maintained. Generally, salt problems from
irrigated agriculture can be most effectively dealt with by increasing the
efficiency of irrigation.
In summary, consideration of the physical and chemical characteristics
of pollutants is the basis for the design of control measures. An effective
practice is one which prevents the delivery of a pollutant to a water body
by controlling the availability of a pollutant in the field, its detachment,
or its transport to a receiving body of water. The extent to which a mea-
sure controls a pollutant pathway will often be dependent on physical vari-
ables such as precipitation, soil type and topography.
19
-------
SECTION 3
NONPOINT SOURCE CONTROL PRACTICES
CRITERIA FOR GROUPING NONPOINT SOURCE CONTROLS
The control of nonpoint source pollution ultimately involves the selec-
tion and design of practices on a site-specific basis. Before this is done,
however, candidate practices must be identified and screened. Candidate
BMPs are grouped in this section according to pathway control mechanisms and
practice permanence to serve as an aid in preliminary design stages. Both
of these grouping criteria are compatible with NPS evaluation and analysis
methods described in Section 4.
Source and Pathway Control Mechanisms
Pollutants may be controlled at their source or point where they become
available, during detachment, or during transport. Source control of pollu-
tants, if practical, can be a very efficient form of control. Source con-
trol is the management of a potential pollutant when placing it in the field
environment. Crop nutrient and pesticide management practices are examples
of practical and efficient source controls.
Control mechanisms which act during detachment and transport are also
effective in reducing pollutant delivery. Practices which control the soil
eroision and sedimentation process will be effective in controlling pollu-
tants that are strongly adsorbed to soil. Practices which control overland
flow and subsurface flow will be effective in controlling weakly adsorbed
and non-adsorbed substances. Examples of pathway control practices are con-
tour farming and conservation tillage.
Practice Permanence
Given the lack of water quality monitoring data, data quantifying
edge-of-field losses, and data quantifying losses between field and stream
or lake, controls considered should be conservative. Nonstructural prac-
tices which are of relatively low cost to landowners and are related to
efficient use of farm resources should be given emphasis in the early stages
of NPS planning and implementation activities. As the NPS program evolves,
land management practices which are more intensive and permanent may be
required. This sequence of practice selection is analagous to the point
source control program where treatment processes have evolved over the past
sixty years from primary to secondary and tertiary treatment levels.
20
-------
Candidate measures can be grouped according to their degree of per-
manence into three types of practices: management, vegetative, and struc-
tural. Management practices involve changes in timing, chemical application
rates, and tillage systems. They usually do not involve separate field
activities.1 Crop rotations and area devoted to each crop remain constant;
only certain farm management decisions are affected. Vegetative practices
involve changes in cropping systems. They generally must be renewed annual-
ly. Structural practices necessitate capital investment and construction
activities. The risks may be high because of this initial investment. Once
implemented, structural practices can affect farm production for substantial
periods of time. However, they also assure more permanent pollution con-
trol.
CANDIDATE MEASURES FOR THE CONTROL OF NONPOINT AGRICULTURAL SOURCES
Soil and water conservation practices (SWCPs) were originally designed
to conserve the land resource rather than to control water pollution. They
represent candidate measures for the control of agricultural nonpoint source
pollutants, since losses of these pollutants are always associated with soil
water movement or erosion. There is a tendency to use SWCPs as best manage-
ment practices since these practices are already familiar to most farm oper-
ators and there is an established institutional framework. This section
will include a discussion of 1) those SWCPs that double as potential water
quality improvement practices, and ?) some additional measures which may be
used to control agricultural nonpcint r.ource pollutants.
The following discussion of candidate measures has been divided into
two sections: those practices relevant to nonirrigated agriculture, and
those concerned with irrigated agriculture. Although certain candidate
measures are applicable to both systems, irrigated agriculture has unique
pollutant control options because water application is controlled. Many
candidate measures applicable to nonirrigated agriculture are ineffective or
unnecessary on irrigated lands. Practices discussed in the two sections are
grouped according to the management, vegetative, and structural control cat-
egories related to practice permanence. Within these categories, practices
are further grouped, although not explicitly, according to affected pollu-
tant pathways.
NONIRRIGATED AGRICULTURE
Precipitation quantity, intensity, and seasonal distribution are signi-
ficant determinants of pollutant loading. Control or manipulation of these
factors are not current options for farmers. Wastewater collection and
treatment are also generally impractical due to the diffuse nature of runoff
flows from nonirrigated fields. Table 2 lists some general candidate
control measures for nonirrigated agriculture according to the three cate-
gories: management, vegetative, and structural.
^•Exceptions are split applications of fertilizers and pesticides.
2The exceptions are, livestock waste collection and storage systems.
21
-------
£
g
1 >> 1 01
i — (J 4- QJ 1/1
3 -r- l/> *f- I/I 1/1
O •— Q) 1, (D O
t/) O. l/> *r-
"O i-
4-> 1- TJ TJ (J
TO Z) C T3 •—
I- C TO C Z) T3
§(O O Ol
i/> .c 4->
O 4-> C t/) TO
•»- t. c o a.
4-> o o; •*- (/)•!-
fO H- -r- 4-> C O
O i- TO O •'-
• — C Z> •<- 4-> C
CL+J -r- U
(O C T) 4-> •!- T3
ZJ QJ TO r- C
i- O N •— Q. TO
•r- TO TO C
r- 1- - QJ O
•r- */> QJ 4- -O •«-
+-> -O c: 4- -.- 4->
i- QJ -t— O O TO
QJ a> E c T- . —
H- C ZJ 4-> 3
r— 1- i/l Q.
r- Q.T- QJ O
tD O O "Q. Q.
r- 1_ (/I CT1
DO C • 4-»
1- *••- C W
QJ .C in JZ O •*»
E u c u T- ex
5 TO •r- i TO. c:
C_J E -M •— O O
o
4->
a>
E
4->
5
TD
'zi
o
JZ
(/t
in
o
4J
(O
^1
a.
CL
-t->
c
QJ
s_
13
c
CL.
1_
CJ
en
c
M- C
•r- C
i — fO
•*-> 'o.
u
01 C
in a>
c .c
s.
• o
a> 4-
ru -a
CL-4->
3 C
=3 I
CL O E
O T
O- -r
in in t
a> a> •••
S- r— +
0 >> <
•u u t
(O
u
u
E
o
o
».
ai
n
fO
E
n
- **-
-» c
D O
•!->
3. i.
i- O
D Q-
1_
I) O
D U
C
J ••-
J r—
IJ O
X l/l
4-
O> O
t/}
•*->
"O TO
u
•r- Q.
5 CL
C Ul
TO at
at o
4- a.
i
i.
QJ
CL
OJ
•4-1
"O
c
TO
T3
C
J
S^
o
c
3
O
(J
TO
T3
O C
JZ O
V) •!-
I/I •!-
ai -o
•o c
•i- O
0 U
4-> QJ
a> zi
Q.-I-J
,
o
t/>
en
c
TO
OJ
s_
o
c
T3
C
T3
O
5^
CL
J
O
TO
4-
cn
c
Nl
E
f
i
i_)
4J QJ
QJ •,—
CL+J
U
"D (O
C t-
TO CL
at
i- QJ
c
4-J
C S-
O O
<- CX
^> U
TO -r-
t_
O O)
CL t,
1- rt)
o
C QJ
••- -o
4-> i-
m "O O C
rt) • — \ +
QJ 3 "O i
i— O c: c
QJ "in" re +
-C QJ i
•*-> c; • — a
O -Q -+-
^ -r- Z» n
4-> 4-1 , — q
•«~ TO O '
J -o t/> c
TO
o; t- " o
"O CD -*-* J
•"— Ol C: -I
u -a QJ
•i— +j (
Ln 4_> ..- n
CLr- t.
QJ 3 CX 1
-C O" T
4-> - r
" QJ TO •>-
QJ 4-> t- 4-
i — TO OJ L
-O i C I
i/> O t
O 4- C -r
Cu >— • >
i — C
1- (T3 * 4-
ai i— T3
> 4-> Ctt -
at c oo r-
C OJ 3 -C
JT o aj -r
3: 0.-Q -C
J
j
D "O
1. 01
J TO
n Oi
U 5-
j U
3 C
U <~
5i QJ
>
» TO
: . -^
j >.
4J -C
U -^ tJ
3 TO -C
cr
n t/>
us- a>
3 QJ -f~
- 4J 4->
~ 3 r"
j * j.
1 QJ "O
V C1-
J cn O
- a) !-
<. TD ^
-> O <-n
4^ C
^1 "*~
- ,— -t-1
r TO c
ri r— TO
- 4_> i —
: c o.
QJ
Z)
-Q
QJ
*"
in
ai
in
• -
a>
in
c
o
4-> -
OJ
U
TO
in
in i )
QJ . :
TO
E
4->
a.
o
4-*
TO -t-1
in Q
CL D
O
i_ 4-
<-> O
C •—
•!- U
•+-> >1
at
1- 4-
^3 f—
-n C
? o
C
t> T3
O)
°>!5
C j}
4-J I/I
C QJ
QJ O3
> U
££
CL Q.
E Q.
*.l
-O +3
o -o o
I_i— -r-
tQJ •I-J
r- TO
C 4->
•i- in
in at
TO 4-> in
i in a>
4J r- -,-
o 1/1 4->
fO O
-C. t-
U 4->
QJ c
•i- a» •*->
in a> QJ
^33:
22
-------
t
^^
£•5
3 cr
•(-> V
X <-
i-o
"Q.
CT> D.
O) 3
p — in
i QJ C
in .5 C
in T
(O r-
t. r—
^'5.
m c
(1)
•r- CO
o
X> 1-
o •*-»
c •
+j as .
V 0
CT>
c
CL
CL
O
L. i-
3 O"-
O
-P CL +
C 'f C
0 1- C
O -M C.
lO^-
ID
1
ro
•L.
(O •—
•— C
.c ro
CT O
•(- -i- in
-d r- tO •
CL. QJ C
C
r- 0 O
r- •«- L) >%
SS-oS
«k°^
•*-» fc- 3
QJ -C 3 C
O O •*—
Q. +J C
•o c o
4- d O •*-
S-
Q)
>
O
O
.M QJ
C >
QJ ••-
C •»->
m m
J 6 4J
r C QJ
3 QJ Ol
J CL, QJ
-> >
l£>
1
>> •*->
r— C Cn-D QJ
QJ QJ OJ C C ^C
4-> in JT T3 ••- «J -40
m « h- c i-
3 QJ QJ -C C
CT C • D- 4-> -M -r-
QJ O t7> 2
•o c a> ••- o QJ o
QJ 3 r- JZ 0
^ i— O 4J
O -4-> QJ -i- 4- T3 C
i_ «J -f- 4_) O C (13
+j ^ <+- o ro i —
C +•» tJ J= CL
O -M t- ^-> >»
CU QJ O. O
Q «r~ T3 in ^ fO
& -M O -C X» O14J
t! *" ^ i5 Q.f^
ifl in CL4— O 13
in o >»+-> 3
m to QJ -t-> O"
+j i- ^ m -i— o>
C en in (j -C T3 ITJ
(D C OJ O •!-> C QJ
4-» " ro C i — *D L.
3 QJ U QJ QJ « > m >^
r— i. tn -i- QJ 4-> T3
CLOQJO-C(TJ4Jin
(/i +-> QJ +J c in
4- i — 4- "O
nJ l/>
r>-
1
Qi r^
i — in o
•r- -r- O,
u_
tn -t^ *o
in
fD "(/)
L U QJ
QJ 3
01 J 0
o o
t- &-
CL QJ £
•D O
QJ S- C
> O 4-
Em t.
m TD QJ
QJ QJ +-»
4-> U_ 3
4-
o -o
•*-* C >.
i- i —
r^ Q. t-
-•--(- OJ -
•r- -M QJ 0
4-J QJ C
to I- o
QJ ,—
4- CTi fO
4- C L
30 n
O
O 3 C
**/i 1^ t- C
QJ in O
QJ +-> 1- QJ •—
•^ 2* 0 § C
•*-> ••-> in o
U 4- C
QJ 4- O I -O
4- O U O C
4- C •— <0
QJ 3 X) f—
V C 10 W>
XJ TJ • —
C T3 •— O -r-
QJ CL^: >-,
CL +J r— O-
O "O +-> QJ
»— C J JZ QJ
in *o O ore
QJ T: •— •— c
x: c o m o
+J (O 4-
i— "D >>
QJ >> QJ r—
Ol QJ QJ "O C
C -C -C (D O
(tJ •*->•»-> W
.C Ol QJ
U 4- >> .—
O i— QJ -Q
O CL S^ ro •
-M U •«- QJ
in OJ U o Q.I —
in
QJ
U
(O
L
L
QJ 4-
•<- O
4-
-C
fO -4->
01
QJ C
O r~
-D
rO OJ
O.
c • —
•t- I/I
4->
m QJ
c >
t- U
L. OJ
O 4-
4-
4- QJ
4-
O CD
C .E
3 -M
S_
in
in C
S- +J T3
•^ j: -^
in
c
TJ
"O 1-
c o
ft}
c
in o
C -t-
o +-*
••- CL
m QJ
t. 'J
QJ '-
> QJ
O C
CD
CM
1
23
-------
4_ (— M-. r—
-(-> 13 -M
*T3 f -- ITS r
U t—
§-r- O M
-r- D-M
5- nj
O -i-
OJ U
OJ O OJ C OJ
9/1
L.
-------
Management Controls
Reducing Excessive Chemical Application Rates (NIA-1)--
If fertilizer is not a major production cost relative to other inputs,
it is often applied in excess to insure maximum yields. Fertilizer rates
however, should be based more closely on crop needs, taking into account the
residual nitrogen content of the soil, past nitrogen additions (from
legumes, manures, and commercial fertilizer applications) and prior and
anticipated nitrogen removal (to crops, in deep percolation, through deni-
trification, or with eroding soil). If all these components are considered
in a nutrient budget, a fair estimation of present nitrogen needs can be
made. Manure and manure handling systems directly affect the nutrient (and
especially nitrogen) content of wastes, and therefore manurial nutrient con-
tent should be estimated before developing a land application schedule.
Pesticide requirements vary from year to year, depending on climate,
insect and weed populations, and other factors. Consequently, application
rates of pesticides may at times also exceed what is needed, particularly
for high value crops where financial losses can be large if pests are not
controlled.
Timing of Application (NIA-2)--
Applying nutrients and pesticides at times when they are most needed by
the crop can be very effective in reducing pollutant losses. Fall fertili-
zer applications of nitrogen can be completely leached below the crop root
zone before the growing season begins in areas of considerable winter pre-
cipitation. Early spring applications can also result in appreciable losses
if heavy rainfall follows. The availability of machinery and labor has a
marked influence on fertilizer application schedules. Also, crop cover may
limit fertilizer applications during the growing season. Fall applications,
where necessary, should utilize ammoniacal (NH^) nitrogen to minimize leach-
ing, and should be applied after the soil has cooled to less than approxi-
mately 50°F to prevent microbial conversion to the mobile nitrate formed
during warm fall weather.
Pesticides should be applied to achieve maximum effectiveness and mini-
mum loss. Insecticides, for example, are more effective at particular
stages of a pest's life cycle and applications can be timed accordingly. In
areas where pesticides are used extensively, the timing of applications can
be linked to weather forecasting. Critical periods when even extremely low
concentrations can influence fish spawning should be avoided.
Improved Method of Application (NIA-3)--
The incorporation of manure and fertilizers in the soil reduces
losses. Incorporation can control losses in surface runoff (especially for
early spring applications when rainfall events closely follow fertilizer
application). Incorporation also makes nutrients more available to plants
25
-------
and reduces volatilization losses by nitrogen, thus increasing the effi-
ciency of application. The method of applying pesticides can also strongly
influence potential losses. Pesticide losses through drift and volatiliza-
tion can occur with aerial applications of some pesticide formulations or
with certain types of spray equipment.
Improved Timing of Field Tillage Operations (NIA-4)--
The timing of tillage operations can markedly affect pollutant deli-
very. If the soil is tilled soon after chemicals have been applied, the
quantities of those chemicals lost in surface runoff can be reduced. While
those fields having a low sediment delivery potential can continue to be
fall plowed, those with high sediment delivery potentials should probably be
plowed in the spring, or a form of reduced tillage considered where possi-
ble.
Using Alternative Pesticides (NIA-5)--
The persistence, adsorption characteristics, toxicity, form, and method
of application all interact to determine the effect a pesticide will eventu-
ally have on water quality. In general, persistent, soluble, and/or highly
toxic pesticides have the greatest potential to impact water quality.
By alternating pesticides, the effectiveness of each application can
in some cases be increased, the amount needed reduced, and development of
resistance among the insect or weed population minimized. Biological con-
trol and integrated pest management programs (combinations of biological,
chemical, and cultivation control) have gained popularity in recent years.
Using Insect-and-Disease-Resistant Crop Varieties (NIA-6)--
If a crop is more resistant to disease and insects, it will require
less application of insecticides and herbicides. Although yields may be
somewhat less for such varieties, the decrease in cost of chemicals may
compensate somewhat for the decreased production. These resistant varieties
may be useful in combination with other practices, such as reduced tillage
or zero tillage, which might otherwise require additional pesticides.
Optimizing the Time of Planting (NIA-7)--
Insect or disease damage can sometimes be reduced, and/or pesticide
application rates decreased, if a crop is planted at the appropriate time.
The life cycle of the pest will determine appropriate planting dates in such
situations. In some cases, it is advantageous to plant the crop at the
earliest possible date. This is generally true, for example, in dealing
with many of the plant diseases found in the nation's extensive winter wheat
crop.
Using Mechanical Weed Control Methods (NIA-8)
Employing mechanical methods to control weeds will decrease the need
for herbicides, but labor and machinery costs will be increased. In addi-
26
-------
tion, these methods may damage soil structure. Pulverization of the soil
can lead to decreased infiltration, increased erosion, and increased losses
of sediment and associated chemicals. Creation of "tillage pans" due to
increased compaction can decrease crop root penetration and nutrient uptake
by crops.
Reduced Tillage Systems (NIA-9)--
Conventional tillage can destroy intrinsic soil structure, decrease
infiltration, and increase surface runoff. It also incorporates most crop
residues into the soil, leaving the surface exposed to raindrop impact and
to maximum runoff energy. Reduced tillage systems include a spectrum of
practices, from conventional moldboard plowing with fewer land smoothing
operations to minimum soil disturbance. The effectiveness of these prac-
tices does not depend as much on the tillage operations per se, but rather
on the amount of surface residue remaining on the field following tillage.
The increase in surface residue reduces the loss of sediment, and hence the
loss of soil-associated nutrients and pesticides. It also assists markedly
in wind erosion control. The increased mulch and improved porosity of the
surface associated with reduced tillage, also significantly increases infil-
tration. Deep percolation losses of nitrate and of mobile pesticides may
actually be increased by this practice in some cases. Reduced tillage can
be used for most fields where crop seedbed preparation is needed. Limiting
physical factors in the suitability of reduced tillage systems are climate
and soil drainage. A higher variabilty in crop yield (and, thus higher risk
to the farmer) can be expected than with conventionally tilled fields.
No Tillage systems (NIA-10) —
No tillage (or zero-tillage) is most adaptable to well drained soils in
areas where the length of growing season is relatively long. As with re-
duced tillage, no till is very effective in reducing surface runoff and con-
trolling soil erosion. However, additional pesticide applications may be
needed to control weeds or insects which would not be needed with conven-
tional or reduced tillage systems. An even greater variability in crop
yield is evident for no-till planting than for forms of reduced tillage.
Contour Farming (NIA-11)
Tillage and planting operations which follow hillside contours increase
surface storage and water, increase infiltration and decrease runoff, sedi-
ment, and total pollutant losses. These practices are generally not imple-
mented on poorly drained soils, and are difficult to implement if the topo-
graphy includes complex slopes. They have little effect on level land, on
slopes greater than 12%, or on slopes longer than 130 meters. Intense
storms may break across the contoured rows and greatly reduce the effec-
tiveness of contouring. Contouring may also be more difficult when wide
equipment is employed.
If tillage and planting operations are performed across the field slope
but not on the contour, it is sometimes referred to as graded rows. The
slope of the grade is usually less than 1 percent. The effects on water
27
-------
conservation, soil erosion and control of pollutant pathways are similar to
contour farming. Although graded rows would affect pollutant movement less
than contouring, where soil drainage is a problem this practice acts as a
compromise measure.
Vegetative Controls
Meadowless Rotations (NIA-12)--
Crop rotations can be effective in reducing the need for pesticides,
since insect population cycles will be disrupted if host crops are elimin-
ated from the local area. In particular, by employing a meadowless rota-
tion, different high value row crops can be continuously grown in an area,
and certain insect pests controlled or eliminated as well. For example, a
corn-soybean rotation is commonly used in areas of the midwest where corn
rootworm persists. The rotation not only breaks the insect life cycle but
the nitrogen-fixing soybean decreases nitrogen requirements for the subse-
quent corn crop.
Sod-Based Rotations (NIA-13)--
Sod benefits the soil in many ways. It improves soil structure, which
in turn increases infiltration and decreases surface runoff. It can also
reduce the amount of fertilizer needed for following row crops because
grass-legume sods act as a source of nitrogen.
Winter Cover Crops (NIA-14)--
Crops which cover a field during all or part of the nongrowing season
influence pollutant loss in two ways. First, increased crop cover and sur-
face residue decrease soil erosion by water and wind. Second, a well estab-
lished winter cover crop can take up nutrients which would otherwise be
lossed via overland flow, erosion (from rainfall and snowmelt events) or
leaching. The main disadvantage of the practice is that the winter cover
crop usually must be seeded in the early fall, amd plowed under or chemi-
cally killed in the spring. This requires more labor, fuel, seeds and
chemicals for the overall farm operation.
Contour Strip Cropping (NIA-15)--
If greater effectiveness is needed than can be achieved by contour
cropping alone, alternating contour strips of row crops with strips of sod
can further reduce soil losses and associated pollutants. This practice is
particularly useful on long slopes or in areas of particularly erosive rain-
fall. In these situations, surface flow can accumulate enough energy to
render normal contours ineffective. By reducing the rate of surface flow
and increasing infiltration, soil detachment is decreased and surface stor-
age of water is increased. In sod-covered areas, the soil is protected from
raindrop impact, so fewer soil particles are detached from the surface.
Soil structure is also improved, and nitrogen is added to the soil if a
legume is included as part of the mixture.
28
-------
Permanent Vegetative Cover (NIA-16)--
If the potential for pollutant loss from a particular field is high and
there is no other cost-effective management practice available, it may be
necessary to leave the field grass-covered on a permanent basis. This pro-
tects and stabilizes the soil, and decreases surface runoff and erosion.
Field Borders and Grass Filter Strips (NIA-17)--
If pollutants are not being controlled adequately at the source, grass
strips can be placed at field edges. These areas function as filters where
sediments and associated pollutants become trapped. Potential pollutants
can then either decay, infiltrate the soil and/or be taken up during plant
growth. In order for the areas to be effective, they must slow the rate of
run off sufficiently to allow some of the sediment-borne pollutants to be
removed. The effectiveness of a field border or grass filter is dependent
on the velocity and depth of flow entering the sod area, the topography, the
width, and the quantity and quality of plant cover in the grassed area.
Buffer Strips Along Streams (NIA-18)--
Buffer stips along streams serve the same function as field borders,
but are generally farther from the source of pollution. Again, the charac-
teristics of the buffer strip and of the runoff water passing over the buf-
fer strip determine the settling rate of pollutants and hence, the effi-
ciency of the control practice.
Structural Controls
Terraces (NIA-19)--
Terraces can be used to change the effective slope length of the land.
Most terraces are graded to facilitate runoff. Runoff may be routed to
grassed waterways or to ponding areas where it can be discharged through a
tile system (PTO terraces). Although terraces do not significantly reduce
losses of mobile pollutants and may increase infiltratation somewhat, they
effectively reduce erosion losses to less than 10 percent of losses incurred
with up-and-down hill fanning. Terraces primarily reduce soil loss and
associated pollutants by decreasing runoff velocities. Terraces are gener-
ally appropriate on fairly deep soils and long slopes of less than 12 per-
cent. Their initial cost is high.
Diversions and Interceptions Drains (NIA-20)--
Runoff originating above a field can often be diverted away from that
field. Such diversions shorten the effective slope length of the field,
decreasing soil detachment caused by surface flow. Any pollutant reaching a
diversion ditch, however, is also more likely to reach a stream. Diversions
can be useful for protecting fields that are highly erosive or that are par-
ticularly hazardous because of high chemical application rates. Subsurface
diversions (interception drains) can prevent the movement of subsurface
flows.
29
-------
Grassed Waterways (NIA-21)--
Stabilizing channels with sod can greatly reduce the amount of erosion
occuring within the channels. This practice is commonly used where channels
remove runoff from contoured or graded rows, or from terraced channels.
Subsurface Drainage (NIA-22)--
The principal water quality benefit of subsurface drainage systems is
increased soil water capacity. By lowering the water table and increasing
percolation, lower antecedent moisture conditions can reduce runoff. For
soils with low permeability, or during high intensity storm events, however,
the influence of tile drains on runoff is greatly diminished.
For repeatedly wet soils, subsurface drainage may improve plant growth,
which in turn may reduce surface runoff and erosion. The net result, how-
ever, may be an increase in nitrate losses from the area due to a reduction
in denitrification associated with anaerobic conditions.
Detention Ponds (NIA-23)--
Detention ponds collect runoff from agricultural fields. They enable
sediments, and pollutants adsorbed to the sediments to settle out of solu-
tion. Available phosphorus and nitrogen can then be taken up by aquatic
plants and non-persistent pesticides will have an opportunity to degrade
before the water is released.
IRRIGATED AGRICULTURE
In arid regions, the timing and intensity of water applications are
determined by the irrigation method. Deep percolation and runoff losses can
be reduced by controlling various aspects of the irrigation system. Table
3 lists some of the measures that can be used to decrease pollutant loading
from irrigated fields.
Many of the soil and water conservation practices for nonirrigated
agriculture listed in Table 2 are not compatible with irrigated agricul-
ture. For example, grassed waterways are not utilized in arid regions
because of the water demand associated with them, and because of the removal
of valuable irrigated land from production.
Current legal and institutional mechanisms for allocating water between
irrigation districts and farms are responsible for much of the inefficiency
of irrigation, and associated pollution. Institutional changes should thus
be viewed as primary control mechanisms. Most institutional controls, how-
ever, should generally be considered only relative to a longer planning hor-
izon.
30
-------
_
§
CJ
o
LJ
S
o> r-
°?
CJ C -r- ••—
c -r- re
•r- l/> O
O I/) •»-
CU i—
u u
•r- -O •*-
E QJ c: > QJ c
S^ O O -C O
c_j -r- S_ (_) .1-
•i— i—i QJ re
31
-------
c
o
4-J
ro
O) CD
N •!-
E i-
4J *~
O.4J
O C
01
O ••-
4-> (J
-0 «4-
Ol <+-
i- QJ
3 1-
CT O
OJ H-
s-
•a
O) C
.0 »f—
E >
r^ ^ -°
•i- 4-> m
4J -r- 0
C •!-
Q_.r- ^_
O) .— a.
0 oo ro
QJ
O 1
•r- I-
o at
-C CL
LJ O
OJ
4-> X) QJ
Q. CDi—
2.5 p
S.S,-
O
4->
C
s~
ai
rsj
4->
1-
O)
H-
Ol •
4->
rt3
1-
0
CL
a
o
c
o
O!
£><
i — C
at £
i/i
c
o
-M
-o -
T3 C i/i
QJ IQ at
4-> (0
(0 -0 to
.^S°
i. ZJ
i. ~o at
•F- at 10
en ai
X jc.
O • 4J
i i. */i
J- a* at
3 fcO u
M- i/J 3
C 1^"°
o *"~ t^
4_>
c c c
O Ol ra
4-> ••-
fcO (O -O 10
ai s- aj E
J to QJ to a*
O O to to
r— ,
at to vt
C CD (O
jo to 01 at
E -r- ,— (^ CD
3 4-> r- U (0
. O -i-
- a. ^: T3 4->
- TJ U C 1
D > m m o
rt Ol LU •— C
to
ai
QJ to
CD >^
ra I/)
QJ E T-
(J QJ |—
1 s.i
QJ
E >vt-»
**- ro o
^ QJ e
i+~ Of~
QJ4->.,-
•— O
L- (/>
QJ C
zi -i- cr
-D i- c
tO Q. C
OJ 3 T-
i«- IO ffl
r- tO C
0-r- -^
>- 10
4J QJ p
C1 CT>
C ••- 4-J
•r- 1 Cn
to o -r-
0 C i-
S- L.
O
c -o c
•r- QJ O1
1 U 3
^ cr
i- ~a QJ
o at i~
LL. i- «+-
to
a>
c
•o
13
O
4-J
C
O
C-J
3
&
u
at
JD
T3
13
O
.c
c
o
(O
(J
o
m
4->
5
en
c
E
c
E
TJ
O
QJ
Q.
fO
01
>
00
u
X
a>
4-J
o
c
-a
1
a>
-o
u
0
c
QJ
O-
O
4->
-o
QJ
O
rO
e
o
c
c
c.
a.
•a
c
CD
d
=3
~TD
QJ
^:
u
s_
QJ
Ol
jQ
Ol
^->
•t^
U
<4-
scheduling
to
QJ
e
X
e
en
C
at
-o
O
QJ
4-J
•a
c:
i
OJ
-o
c
o
at
i/i
at
CD
S_
o
c
-C
QJ
>
O
^
-Q CTi
r- C
-Q -r-
•r- C
X C
QJ tJ
M- Q.
QJ
M
Ol
O
>
QJ
CT
C
-o
Ol
u
0
4->
CD
S-
«
1 moisture
o
-C
4->
C
o
4J
0
c
(J
u*
o
'o
1_
o
at
4->
at
E
to
c
o
4-1
cn
V-
a>
i_
4_)
O
00
Ol
>
at
c
»+-
C
at
0
4->
QJ
a.
0
t.
at
to
to
^
4->
O
4->
o>
1
5
0
^
Oi ra i- 3 cn
> £ ai c
o a. x> -^
i- t, c s-
CL at F—
-------
(O
i/> 13 QJ •+->
S C *-> «>
o ro ^ "D •
<— ~o cn o> QJ
r- cn QJ c cn i-
.."S
(/) ,—
a
en
- cn ,w —
•> o ~o QJ
(/> I- •»- O <4-
O 4-> O O TD
CL c ro i/> c
QJ o CL4-> i- ro
-O O ro C -r-
O QJ ro C:
4-> > CL O
r— ,— r— QJ QJ •-
•i— O ro 1- S- (/»
ui S- c CL O
•*-» ro QJ
4- C O "O -C
o o c t—
U */> ro
QJ C
QJ I- •(- C
s- QJ ro o QJ
4- 4-> 4-> -r- 1-
ro c +J ^
i/> j -f- ro 4->
r- ro s- u
ro -o E •»- ^
c: QJ ex s-
ro > c: »/i -»->
u o o c: w
s- -i- ro
tr> Q. -•-> t- .—
c E ro +J o
.,_ .,- 40 O t_
CL QJ CL4->
QJ i- cn ro c
QJ O QJ > O
b^ 4- > QJ U
E •
c:
QJ O
QJ ro
LO QJ
~i
4- -O
O QJ
-O
QJ QJ
QJ O
s- s-
ro U) C C
O O
en c •!- T-
y ^) O QJ •-
"D 1~
-o ro "o i-
C "D QJ C *r-
ro i— crt ••-
QJ S C
t\J -i- cn
- c' g wi
•i- O ^- >,
ro t. i— -r- wi
1-3 i/> QJ 4-
C. E
O Ul U> -i~ C
O t/> i— -4-) t-
QJ
-C QJ QJ
u x: .c >,
• r- 4-> *-> -Q
^
S 4- -»-> c:
O O O
i/i d> -r-
CL 4-> *-> ^
O 1- O O
i_ ro i- *-
O Ct CL QJ
L^; OJ
> o E TJ
. 13 -r- OJ
) i/> -a +->
QJ c
) t/l l/l ro
• a f—
O 0 Q
cn 4-> X3
u c
QJ QJ ro QJ
. r- ,_ (U r- -r- r
T 4-1 JD QJ
- T3 - 34-
J C C O 4-
3 ro O "O QJ
"O
3i^ " C
i -Q
• ro ro QJ
J 4-J 4- 5 ^
- O •— O O
"O C <—
c c: 4-» c
o ro c ••-
o x: QJ ro
o E z:
i- cn i-
d ro 3
E c 4-»
-^ ro QJ
O 4-> CL
1- CO
<_> -r- ^
33
-------
4-> t-
rB •>- E 0*
cr> u 4-» E
C QJ 3 0)
£ 4- ^ o
tO QJ O +•*
c a.
cn-c »r-
O V) > -r-
E a» i—
4- QJ J
4- 4-> o ai
O I- I fc.
C 0 0
i. (/) "D
>— 0)
4-> 4-* QJ L-
U ••- T- <0
a> 4-
i- ~a «/>
-o > i/>
C <0 4- t.
O X O OJ
•<- ro > •
t/) -C •- E
J- T3 4J -O ro
QJ •-•• C7> QJ
VI
E
ro
O Q-
•r- QJ
i/l U
S- L-
QJ QJ
•r- C
Q >-«
O
CSJ
^
i- E C
-O *-> -r-
S- (0
S-
t- -o >>
ZJ QJ 4->
4- r- -r-
1- > E
0 dJ O
•4-1—4-
"0 I/) C
C -r- 3
CO
•— 4-> C •
•r- O -O
(U ••- Q>
O. CTi QJ
L. C (- 0
Q- O J- C
O 4->
4-> rO * O>
CTl QJ ^J
s'Els
ai
c
QJ
>
QJ
__»
•a
E
«O
_J
CM
X
> 4->
C O
O E
o o •
4-» a. c
S- QJ
SSJ
C T3
O S- QJ
CJ QJ CO
O I- "O
•a •»- o
QJ L.
01 (/) QJ
3 OJ
Q.4-
QJ O O
i-^ ^_
W C
C O
(0 (D ••-
U > 4J
«J -.-
rjl O W1
-^00.
OJ -O
> O
Q> +-* "O
• QJ '1—
o 0.4-
t/l O I
i — i — cr
QJ
4- yi t- TJ 4J •
4-* ra 3 QJ "O
(/I E*rO O"OX)wiQJ
r— E CU ]C •*-> QJl — C^.C
(/l^1*- (/) C l/i '1— (/)«/> l-O
QJ i- 01 QJ co I-EOro
4->QJ QJ E i/l QJ 3QJ -i— C QJ */>
(Q CL T3 r— LJ (/I > U) *o QJi/iQJ
i-t- t- C 1_"O QJt-QJ •!— 3
{J Q. QJ QJ QJ QJ Q_>c >!_.—
•i- i- t- »• .,- O-CD4-1 QJ T»
C>>4-> £ OJ ^ -^* QJ >i— t-rtjj 1 • r~
OJE C1*-*-1 T3 QJ *-»J S OJ'*-1*
P— roQJQJrO •— C'OO1'—
£ -Q J •• J3i/> tozjl-
QJU -r- ro 4— CT •(— 4~> t-OQJO i-*O
> Q.r— 4- E ro Z) •»— U 1- QJQJi/i
SO rOO C3E4-1 QJ
a. c i- —
O 3 O
C O i/> V
O rO 4->
•i- QJ QJ C
4J CT) E O
C ro t_J
QJ CL J
4-> QJ 0-0
QJ QJ t— C
a: t/i ii- /o
csj m ^f
OJ OJ OJ
34
-------
I
C
0 Q. >> 0>
U
irt >>••- -0
•r- 1_ 3 C
O> O* *O
(U >
l3^ " °
o-o 2 t.
> 03
t. r- 4-
204-
O Of
»— -C
.C
•O O>
IO 01 .C
C7> VI
C 3 O
.,- -r-j+J
vi -o
0)
£_>>.£ u
V)
*/»
£ 0)
t— v>
O) t-
2 O
O 4-J
4- •>
= C
CD O
C i/>
iii
o
V) S-
Q 4->
O
O
4J en
C
= o
(j 15
5 .
3 C
="-§
•<- C
0) *r-
(J
•o o
1- 4-»
-D E
C
IO VI
c .c
0 4->
IO (~
u at
•i- >
O- O
(O 31
4- •
O 4-
4-
O t-
•r- 4J
3 2
VI
l/l
O
U
C
ai
u
4-
«
0)
t-
^_
0)
.a
fO
o *
•r- O>
a. o
Q-r-
IO t/1
c
o
a>
JT "O
40 C
to
4-
O V)
V) 0)
I- t/)
3 >»
£ ^
4- U
O) 5
£ +->
W3
V)
ai -o
c c
•r- IO
1?
(J &-
s-
T3 3
l|
•r- Q.
0)
a.
3
O •
C ,—
IO •!-
"O IO
ai T—
r- tO
o >
i- tO
C 1-
O 0)
U 4-1
IO
>> 2
1o c
U IO
4J §
O C
+-> O
3
O) -f
-O 5
4-
0
£
3
fO
0>
4-
o!
C
f1"
o
o
-C
a>
E
1—
C
J
4-> Q c
t. CL
O 3
^ VI
VI
t-
•i- 1C
B c '
V) I/I ^
>» 3
>,
Ic
1- 3
5s-
4-
•o o
O 4-1 t
a. c c
E OJ -r
s
i
J
r ?-
i 4.
'
'
c
'
Oj 1
4V <
§t
-3 4
i. >
- 4)
j !B
D ia
- OJ
3 c
- ai
-• CL
E
4"
C
^f
W
4->
£
(J
ro
4-
o
o
at s-
rjo at
4-> 4-1
4-
o
c
o
O
£ S
A)
IO
(U
*O 4J O
t: c u
t) C
«,°u
•— 4- O
•^ 4-
< It3 S
S J- In
• — 13 >,
5S,
35
-------
TJ
4V
C
3
fO
LU
s
4->
U
T3
a
£
4->
C
4-1
3
r—
o
a.
c
o
4->
a.
t-
o
i/i
s
asure
QJ
^
QJ
4V
tJ
•i-
•o
C
re
o
•o
QJ
(/>
.c
4V
a.
o
o
CO
in
re
Ul
1/1
OJ
0.
s
QJ
Q.
<
1
>0 1-
4-> 4->
•r- QJ
*— t3 -Q
re QJ
3 i/> C
CT 3 re
U
-O QJ
O J3 i-
O QJ
Cn c +J
re
Sin 4V fo
1- Cn
en o QJ ••-
O •*-* L.
QJ4V 5 i_
54V *'^
O QJ I-
U- C .C O
*-* QJ
E en
QJ O3
4-> C
S_ l/l T-
QJ >> re
-M l/> J_
re o
X OJ
- a E
m QJ 3
K- or a
r^.
OJ
i
<
i
*
4J
>,
H-S
5 c
^S
-o »
o •—
en o
in
4-
O QJ
,c
4-> 4->
c E
(/i p
*~ 4-
L.
aj -o
w^
S|
•o o,
c
^ en
o =
en QJ •
-Q T3
QJ QJ
J= I~ 4->
4V OJ
*»- M- i-
•— re 4->
i i
i i
i i
i i
i i
in o >> r-T
, — 4J 4V r— |
•i- -r- O ^^
O ~O E t. C 4->
in c C i -r- c
QJ O QJ t. QJ
CL 4- TD a.—
QJ QJ -r- -i- I/) O
QJ C C (/)••-
T3 O 13 4J **-
. . O) k|_
i — re > QJ i
QJ p -O T3 QJ
> TJ C 3 •.- t-
•r- QJ O. ,— r- O
•*-> -^ E O O E
re *i— ••— c tn
i— -O -r- QJ
QJ ^ « T3 t-
l_ -Q C C C C rt3
c re o o re
•o re •-- ••- in c
C M- 4-> 4J QJ E O
re-oM-re recS-r-
QJ O O en-.- 4J 4->
w ,— c -r- •«- i — tn re
Q. > S 'o. u i ui*-?
re QJ a. -r- o t.
U »— QJ re 4-t L- t.
W U 4- QJ -r-
•o QJ re c o "i —
C -O H- O 4-> .* QJ
^ C 3 +J § > -^ m
ra in re -M T- t- *4—
•MO en m o. CL i-
(O 4-> •!- >, t tO 13
• — tn c t- in i- tn
U- T3 QJ i- QJ .
r— > -i— I/) 4V l/> C
I- QJ QJ -i- C 1- re
O-r-I-V- -C QJ QJ _C
LL.4-Q.O 1— U^-4->
i-
0
(/)
a ^
o c
i- ••- C
i- m i- o
3 re QJ -.-
LJ_ OQ i— 4V
w f«
r- T3 C Cn
QJ QJ •(-•!-
QJ •*- Q. i-
I CT1 tO n
CO Cf>
C\J OJ
I i
(/) *O
1/1 t_ C
Q> t! > ;i .
*3 re 0' re
o i wi
^ i _ s *" S •
-g*5 ° S0.
j- o'S "- B^
^SS | gg
•— C C • 43 4J Jn
QJ -r- re QJ m ^
"° ' — "~~ ^ i- i. to
Q. 3 Q .,_
"* -^ "M QJ 1 1 ,rt -
•r- 0. 0. 1^ S "^ ^
5|bi t ?«
ra +j^^ ^ ra?
m •*" O -Q i_ 4_> -
in: s r«
iCT; ° J9 Q> o QJ
^> ^ E u
l/l O •!— m U_ -n
r--M Z £ ^^
m re a. IT m
•r- 3 S- 0 3 -a
-c TJ QJ ^ S ^
•M -r- -O >> j^ ^ .^
j= £ o T 3 ° ^
5-5 ° re w S-f
•r C C QJ 4_ .£ ^
3 — '•- c »- TD 0
C C
0 S-
•M 4->
t> QJ
cr -o o;
I^ » (Q 14-
s- co
—• O 4-*
QJ 4-> QJ re
-^ QJ -M O in
0 r- « Q.»
•i- i — QJ in o
1- O 1- -r- I —
t^- (. ) t- r-i LI
o , —
rO m
i i
-------
C d> C HI
O i/l O VI
O Wl U */>
O O
fO r—
O i—
r- Q.
I 4-> QJ
: «3 t/i
1 S 3
i O C
] I/) fr-
) QJ U
: 1/1 cu
- QJ S_
Q. Q
t-
t. 4-> •«-
CT> E -o i- •
wi >^
QJ QJ 3 r-
C C 4->
QJ U
-
QJ
C£. r
^- 'Q 3 L
iD I/I •*->
U -•-> C -O
1/1 o *— QJ '*~
3 JT C7) QJ U
S- Q. c: O T3 -^
O Wl tt^ t- -I— 4->
-C o cn+J o t/i
cxx: p •- •«- -•
o
CX OJ
-g -
QJ O QJ
37
-------
Management Controls
Reducing Excessive Chemical Application Rates (IA-1)--
As discussed in the section on nonirrigated agriculture, tailoring
nutrient applications to crop nutrient needs can minimize nutrient loading
to streams. Pesticide application rates should also match estimated needs
with due consideration given to economic risks and uncertainties.
Improved Timing of Chemical Application (IA-2)--
Fertlllzer application should not be followed by excessive water appli-
cation. This practice assures maximum nutrient uptake and minimum nutrient
losses via deep percolation and surface runoff. Fertilization in fall or
winter should be avoided if possible since large leaching losses of nitrate
can occur with winter rainfall. Irrigation techniques, such as sprinkler or
trickle irrigation, are well suited to tailoring fertilizer applications to
crop needs, for they enable farmers to apply a mixture of fertilizer and
water throughout the growing season.
Improved Method of Chemical Application (IA-3)--
Refer to the nonirrigated section above for a discussion of incorpora-
tion (NIA-3). Application through the sprinkler or trickle irrigation
lines offers considerable promise in certain areas, though fertilization
rates must be modified from traditional levels, and better estimates of
localized salinity effects associated with such practices are needed.
Sprinkler-applied pesticides or fertilizers must not be applied excessively
to portions of the field where sprinkler patterns overlap, and must be
flushed from the lines before the lines are allowed to drain as pressuriza-
tion eases at the end of the sprinkler operation.
Using Alternative Pesticides (IA-4)--
Refer to nonirrigated section above (NIA-5).
Using Insect and Disease-Resistant Crop Varieties (IA-5)--
Besides using crops that are resistant to insects and diseases (as dis-
cussed in the nonirrigated section), salt tolerant crops can also be grown.
This reduces the need for excessive water appliation to remove salts from
the root zone. Total salt and nutrient losses (both from subsurface and
surface flows) will thus be reduced, with only a slight decrease in produc-
tion if any.
Optimizing the Time of Planting (IA-6)--
Time of planting is often more flexible for irrigated than for nonirri-
gated agriculture. Early planting of warm season crops in cold climates
should be avoided if it results in a long period of near-dormancy, or when
added nutrients can leach and/or be eroded from the unprotected soil sur-
face. Refer to the nonirrigated section for additional discussion (NIA-7).
38
-------
Using Mechanical Weed Control Methods (IA-7)--
Refer to the nonirrigated section above (NIA-8).
Appropriate Timing and Choice of Tillage Operations, Reduced Tillage Systems
and No Tillage Systems (IA-8, IA-9, IA-10)--
Proper cultural practices are important if crops are to be grown suc-
cessfully in hot, dry regions. This is especially true under irrigated con-
ditions, where weed and insect pests proliferate. To optimize both salinity
removal from the root zone and efficiency of irrigation applications, deep
tillage may be required to permit greater percolation, greater water storage
capacity, and deeper root penetration into less permeable soil layers. Til-
lage operations also incorporate surface-applied fertilizers into the soil^
reducing volatilization losses and making the fertilizer more available for
plant uptake. This process also tends to decrease nutrient leaching losses
by increasing the degree of contact between nutrients and the soil. Exces-
sive or unnecessary tillage, however, can be detrimental to soil structure
(resulting in crusting, for example) and can increase evaporative losses at
times when crop moisture demand is high.
Each tillage operation on furrow-irrigated fields increases sediment
losses, so decreased tillage of such fields is often desirable. Reduced and
no tillage (zero tillage) systems decrease detachment of soil by raindrop
impact and overland flow. For wind erosion control they are required in
many irrigated areass. The alternative, frequent early-season irrigations
to keep the soil surface moist, leads to excessive erosion and deep percola-
tion of nitrates. Although surface residues are effective in reducing ero-
sion from irrigated fields, some problems in controlling furrow flow rates
and in obtaining adequate uniformity of application may result f^om residue
accumulation in furrows. Thus, the advantages of water conservation and
sediment control may be offset somewhat by decreased efficiency in irriga-
tion water application.
Contour Farming (IA-11)--
Contouring is generally not applicable in irrigated areas, although
contour irrigation is frequently proposed in furrow-irrigated areas with
steeply sloping fields. Limitations with respect to field size, turn-around
areas, and compatibility with existing equipment, are even more stringent
for irrigated than for nonirrigated areas. Refer to the nonirrigated sec-
tion for additional discussion (NIA-11).
Improved Water Management (IA-12)--
The predominant form of surface irrigation in the western states is
furrow irrigation. Historically, most irrigating was done on the basis of
fixed rotation or visual crop symptoms rather than by scheduling the irriga-
tion to meet predicted crop demands. Where an ample supply Of water exist-
ed, many of these systems operated inefficiently. Often farmers have used a
relatively small stream size in order to minimize visual runoff losses. As
39
-------
a consequence, they induced relatively large deep percolation losses because
of nonuniform water application between head and tail portions of the field.
Uniformity of furrow irrigation is maximized when the "intake opportun-
ity time" at both ends of the field are equal. Since water is conveyed from
the head end to the tail end of the field, equal intake opportunity times
along the furrow are not possible. The least watered area of the field is
at the bottom end, so under existing management practices, the highest
attainable irrigation efficiency is achieved when the root zone in this area
is just refilled.
There are two "operational" wastes which occur under furrow irriga-
tion. The first is the water percolating below the root zone; the second is
the runoff from the lower end of the field. Efforts to reduce one often
increase the other. Salinity control practices which reduce deep percola-
tion, for example, will often cause high runoff to control one type of waste
without excessively increasing the other may thus require some system modi-
fication.
Call Period or On-Demand Watering Ordering (IA-12)--
To optimize crop production while minimizing water quality impact,
water allocation systems must provide sufficient, but not excessive,
quantities of water to farmers upon request. Insufficient amounts reduce
crop production, whereas excessive amounts can increase the water quality
impact of return flows. The use of a call period, a minimum length of time
prior to the next irrigation during which an irrigator could place an order
for water with the canal operator, would improve water deliveries. Call
periods allow for better scheduling by canal operators and encourage farmers
to carefully plan their irrigations.
Although on-demand scheduling provides maximum flexibility of irriga-
tion scheduling, it encourages less planning of irrigation operations by the
grower. Such scheduling is desirable when center-pivot irrigation systems
are established on irrigated lands designed for surface water deliveries.
Otherwise, unused irrigation water can rapidly fill storage ponds and lead
to substantial erosion on its way to natural or artificial drainage ways.
This is particularly a problem because center pivot systems are normally
designed to begin and complete revolutions at a different time each day (to
avoid systematic crop water stress patterns in the field). Ditch riders, on
the other hand, normally make thier rounds to increase or decrease farm unit
diversions at the same time each day.
Irrigation Scheduling (IA-12b)--
Irrigation scheduling is a well established practice. Irrigation
scheduling services, both public and private, generally combine
meteorological data or evaporation pan information with soil moisture data
to forecast irrigation requirements. Both the appropriate depth of water
application and the proper timing of the irrigations are estimated.
40
-------
Most commercial services also combine pesticide and fertilizer
recommendations with the irrigation schedule. Increases in farm
productivity and reduced pollutant problems are benefits of the service. In
addition, irrigation forcasts allow the farmer to more effectively schedule
farm duties.
Conveyance Channel Maintenance (IA-13)--
Improper canal maintenance causes wide fluctuations in water levels
along the canal length which make it more difficult for farmers to apply
water uniformly. Uniformity of application, as discussed previously, is a
key factor in maximizing the benefits of applied water and minimizing water
quality impacts. Canals must be kept free of silt deposits and protected
from scour to permit free movement of irrigation flows. Aquatic vegetation
in and along canals needs to be controlled periodically to maintain canal
capacity, reduce unnecessary evapotranspiration losses, and prevent clogging
at control structures. Canal banks, damaged by burrowing rodents or breaks,
must be repaired to prevent excess erosion or seepage losses. Structures
employed in canal systems must also be properly maintained against leakage,
settling or general wear. Improper canal maintenance causes wide fluctua-
tions in water levels along the canal length which make it more difficult
for farmers to apply water uniformly.
Improved Management of System Storage (IA-14)--
Measures to improve system storage assist water managers in timing
irrigations to decrease over-application. Although system water storage
provides operational flexibility and more efficient water use, it should be
minimized to keep seepage and evaporation losses as small as possible. If
water storage is needed throughout a river basin, water should be stored
only as high in the basin as practicable, so that water surface area evapor-
ation losses are reduced and flexibility in diversions is increased.
Improved Management of Return Flows (IA-15)--
Seasonally, many irrigation systems have a larger volume of water
available than is necessary to adequately supply crop demands. Instead of
properly cutting back diversions at the river or reservoir in such cases, it
is common to operate canals at capacity at all times with unneeded water
spilled into wasteways. High flows between irrigations are especially true
for canals that are relatively long and have slow response times. This cus-
tom produces an unnecesary soil erosion from unlined canal ways, and high
pollutant concentrations.
Land Retirement (IA-16)--
A large part of irrigated pasture provides only small returns to pro-
ducers. Most of this irrigated pasture land is unsuitable for more inten-
sive crop production. Retirement of this irrigated land from production
would appreciably improve water quality at a relatively small cost.
41
-------
Vegetative Controls
Crop Rotations (IA-17)--
Use of crop rotations permits the use of deep-rooted crops to "scavenge"
residual soil nitrogen remaining from shallow-rooted crops of high nutrient
and water requirements. Rotations also provide greater economic flexibility
when adopting management systems, permit control of many insects and plant
pathogens, and permit periodic establishment of soil conditioning or salt-
tolerant crops in particular problems areas. Refer to the nonirrigated sec-
tion for additional discussion (NIA-12 and 13).
Winter Cover Crops (IA-18)--
Although not traditionally regarded as applicable to irrigated agricul-
ture, this practice is now used with increasing frequency in sandy areas
where winter and spring wind erosion is a serious problem for unprotected
soil, and during early-eseason furrow irrigation when excessive erosion may
occur. The cover crop may also "scavenge" nitrate which leached below the
root zone of sha 11ow-rooted crops the previous season.
Field Borders (IA-19)--
Field borders or filter strips are a common sediment-control measure in
furrow-irrigated settings. This practice is not well-suited to erosive
crops such as potatoes and close-grown crops such as alfalfa. With erosive
crops, the filter strip rapidly becomes covered and ineffective, and with
close-grown crops, the fine sediment which may be eroded is not filtered by
the grassed strips. Double and triple-planted wheat has been used effec-
tively as a filter strip material for furrow-irrigated fields. Refer to the
nonirrigated section for additional discussion (NIA-12 and 13).
Structural Controls
Diversions and Interceptor Drains (IA-20)--
Subsurface diversions by interceptor drains may be used to intercept
groundwater bearing substantial loads of salinity or nitrates. See non-
irrigated section for a discussion of these control practices (NIA-20).
Land Leveling (IA-21)--
Land leveling is commonly used to prepare land for furrow or border
irrigation. Irrigation uniformity can be increased when land is at a con-
stant grade. This practice, however, can have negative impacts on the
soil. Calcium carbonate accumulations, for example may exist at relatively
shallow depths in the native soil, and distribution throughout the soil
would reduce its quality.
42
-------
Retention Ponds (IA-22)--
This control practice is common throughout the irrigated west. It
appears to be one of the most cost-effective practices for sediment control
in furrow-irrigated areas. Refer to the nonirrigated section for addition-
al discussion (NIA-23).
Seepage Control (IA-23)--
Many unlined canals, ditches, laterals, and watercourses traverse long
distances between the point of diversion and the farm. Where soils are well
structured and permeable, seepage losses may be considerable. Tradition-
ally, canals with high seepage loss have been lined with a variety of alter-
native materials, including concrete, asphalt, bentonite, compacted earth,
and plastic. The economic justification of such lining has been based on
the value of water saved. Converting to a closed conduit of concrete,
asbestos-cement or plastic is a relatively costly yet an effective alterna-
tive that offers advantages of less friction, reduced evaporation, better
maintenance of pressure due to gravity, and improved aesthetics. The salt
contribution from conveyance seepage often exceeds that leached from the
irrigated land, thus conveyance linings can frequently decrease salt loading
significantly. Seepage water does pass through recently fertilized soil, so
its nitrate content is generally lower than that of percolation losses from
cropland.
Flow Measurement and Control (IA-24)--
Poor water management leads to application inefficiencies, which reduce
yields and/or impair water quality. Under-application may lead to localized
salinity problems, whileover-application may cause excessive runoff of sedi-
ments and associated pollutants as well as leaching of salts and nitrates.
The purpose of flow measurement and control in irrigation systems is to
ensure an adequate application of water to croplands, while preventing un-
necessary and wasteful diversion which may result in poorer water quality.
In order to control the flow of water in canals or ditches, structures
referred to as checks and drops are used. These structures control the
slope and elevation of water surfaces. They are also critical in dividing
water, as well as distributing water to each field. Other control struc-
tures include culverts and field inlet devices.
Optimizing Furrow Advance Rate (IA-25)--
The flow in surface furrows should be such that the advance time for
the field is about 25% of the total set time, assuming a uniform slope. If
this practice is implemented and strictly adhered to, deep percolation
losses may be greatly reduced. Flow rate and volume can be adjusted to
furrow slope and length to satisfy these criteria.
Cutback Irrigation (IA-25a)--
With this method, the head ditch or delivery pipe is adjusted so that a
43
-------
large "wetting" furrow stream is introduced to quickly advance the flow to
the end of the furrow. The flow is then "cutback" to a "soaking" flow rate
in order to complete the irrigation. Benefits of this practice include
increased uniformity of application, reductions in tailwater runoff, and low
labor costs. This method is applicable, however, only if sufficient cross
slope is available.
Gated Pipe System (IA-25b)--
Gated pipe irrigation systems combine features of both the improved
furrow and the cutback systems. The gated pipe system can in turn be
controlled by a time clock or a master control panel. When coupled with
on-demand water availability, irrigations can be scheduled according to crop
needs. Length of set can be automatically controlled, so that uniform set
times are no longer required. Stream size can be automatically cut back
when the stream reaches the end of the field.
Multi-set Irrigation System (IA-25c)--
The automatic multi-set irrigation system combines features of the
improved furrow system with a shorter length of run, by having several
lateral supply pipes across each irrigated field. Irrigations can be
started automatically and can be scheduled to meet crop needs. Length of
set can be automatically controlled, and the shorter length of run reduces
the period of furrow advance. This in turn increases uniformity and
decreases deep percolation. In addition, a shorter length of run allows for
the use of a smaller stream size, which reduces runoff.
Subsurface Drainage (IA-26)--
Any expansive area upslope from existing irrigated lands may cause
waterlogging of downslope areas. If the underlying strata are sufficiently
permeable, tile drainage is an effective means of lowering the water table
and facilitating salt movement from the root zone. In addition, tile
drainage allows for the collection of surface return flows into a master
drainage system for ease of control and/or treatment.
Pump Drainage (IA-27)--
Pumps have also been used effectively for lowering the water table in
many areas. If the water is not too saline, it can be reused directly, or
after mixing with surface water supplies by discharging the flow back into
laterals. For good quality groundwater supplies, pumping serves the dual
purpose of alleviating waterlogging and providing additional irrigation
water. With poor quality supplies, however, disposal of the pumped water
may constitute a major problem.
Level Furrows or Diked Basins (IA-28)--
For flat landscapes and relatively deep soils, fields can be levelled
and diked at one end to prevent surface runoff, and to improve uniformity of
44
-------
application. With this system, irrigation efficiencies greater than 90%
have been achieved in some instances.
Sprinkler Irrigation (IA-29)--
Sprinkler irrigation, if properly designed, installed and operated,
results in both water quantity and quality benefits. Relatively uniform
water application is generally possible on various types of soils, thereby
minimizing deep percolation losses. Little or no tailwater runoff
results, except when infiltration rates are exceeded on fine-textured
soils. Sprinkler systems commonly used include side-roll, center-pivot,
tow-line and solid-set sprinklers. The last two are frequently used in
orchards and vineyards.
There are disadvantages to sprinkler systems, however: (1) poor quality
irrigation water may cause salinity damage to crops and leave undesirable
deposits or coloring on leaves or fruit, (2) sprinkler-irrigated crops are
more susceptible to certain diseases, and (3) the high capital costs associ-
ated with the sprinkler and the fuel and/or energy costs to run the system
may make sprinkler irrigation economically impractical for certain crops.
Trickle Irrigation (IA-30)--
The concept of trickle irrigation is to provide crop plants with near-
optimal soil mositure by conducting water directly to individual plants
through lines or emitters. In widely spaced crops (especially orchards),
losses are reduced because only a small portion of the soil surface is
wetted and subject to evaporation. Little water is lost via deep percola-
tion, since only the plant's root zone is supplied with water. The only
irrigation return flow is associated with the occasional leaching necessary
to prevent excessive salt buildup in the root zone. Because of continual
leaching of salts and nutrients to the periphery of the wetted area, nitrate
concentrations of these return flows can be exceptionally high. There is no
surface runoff, and little water is consumed by weeds with this system.
Trickle irrigation systems are easily automated, but they usually
require skilled technical assistance for estimating nutrient balances and
fertilizer applications. Irrigation waters should be relatively high
quality. Filtration and chlorination are often required to prevent clogging
of emitters. Emitters are economically practical for widely spaced crops if
the average distance between emitters is more than 3 meters. The capital
cost of trickle irrigation is relatively high. Trickle irrigation systems
are particularly adaptable to perennial high valued crops, such as orchards
and vineyards.
Collection, Treatment and Disposal of Return Flows (IA-31)--
In many cases, subsurface and surface return flows from irrigated lands
may be so brackish that no further use of the water is possible. Such flows
can be collected and either disposed of or treated before they enter receiv-
ing waters. Reverse osmosis and electrodialysis are two desalination pro-
cesses that can be used to reduce salt concentrations in irrigation return
45
-------
flows. Major disposal alternatives include deep well injection and
evaporation ponds. In general, the cost of collection, desalination and
brine disposal for salinity control exceeds the costs required to achieve
the same level of salt reduction through improved irrigation efficiencies.
Institutional Controls
Tax on Farm Inputs --
Charging a farm operator in excess of the market price for fertilizer,
pesticide, or water would theoretically decrease the amount used by encour-
aging more efficient use.
Discharge Regulation --
Since the amount of water, nutrients, and sediment leaving a farm in
surface flow can be monitored, it is possible to estimate pollutant dis-
charges. Based on these estimates, loading limits can be established for
farms or irrigation districts.
Regulating Farm Inputs --
Nutrient and water requirements for maximizing crop yields can be esti-
mated based on agronomic research and experience of local growers. Irriga-
tion and fertilizer applications can be regulated to conform with agronomic-
ally sound recommendations.
46
-------
SECTION 4
METHODS FOR THE EVALUATION AND SELECTION
OF AGRICULTURAL NONPOINT SOURCE CONTROLS
The selection of agricultural nonpoint source (NFS) controls involves a
series of steps and calculations. This section presents methods by which
nonpoint source controls can be evaluated. Application of the methodology
to two case study watersheds is described in Section 5.
The selection process presented in this section involves an evaluation
of two basic factors: 1) NPS practice effectiveness in controlling a par-
ticular pollutant pathway, and 2) farm costs associated with the practices.
The steps outlined in this section generally apply to all agricultural
systems. The seven steps of the BMP selection process are shown in Figure
5.
STEP 1: DESCRIPTION OF WATERSHED
In order to develop and evaluate NPS control strategies, physical,
hydro!ogic, and crop/livestock information will have to be compiled.
Physical data includes maps of land use, topography and soils for the study
area. Available hydrologic data, including rainfall, runoff, and ground-
water records, should be located. The types and extent of agricultural
enterprises in the area should also be noted. The technical appendices
document specific, data needs and collection methods.
STEP 2: PROBLEM IDENTIFICATION
The identification of impaired water uses and associated problem
pollutants, determination of the relative contribution from agricultural
nonpoint sources, and calculation of the extent to which these pollutants
should be controlled are of primary importance. This information
establishes the basis for the implementation of control practices. Although
methods for determining the existence of a pollution problem and the
necessary level of control are beyond the scope of this manual, the general
approach is briefly described below. 3
Impaired water use provides the first indication of problem pollutants
and their sources. Cause and effect relationship must then be established
3The reader is referred to Chapra (1980), Thomann and Segra (1980), Di Toro
(1979), Wineman et_ _al_. (1979), Heaney and Ammon (1979).
47
-------
Figure 5. Steps of the BMP evaluation process,
Step l: Description of Watershed
I
Step 2: Problem Identification
Step 3: Determining Applicable Control Measures
I
Step
4:
Choosing
the
Unit
of
Analysis
i
Step
5:
Establishing
the
Base
Condition
I
Step
6:
Evaluating
the
Control
Measures
I
Step 7- Developing an Optimal Control Strategy
between water uses, water quality indicators, and contributing pollutants
(Figure 6). Water quality monitoring data is in all cases required. The
extent of monitoring will vary considerably with the availability of project
resources. Proper design of limited grab sampling programs can yield
valuable data.
The relative contribution of point and nonpoint sources to pollutant
loads above background or natural levels must be estimated. If the
NPS load is small compared to the point source load, NPS control may have
little impact on water quality. If, however, the NPS load is significant,
then the relative contribution of various nonpoint sources must be
estimated. In evaluating loads from agriculture, a 'manageable1
agricultural load must first be determined; a manageable load is defined
here as the component of total agricultural load. If the manageable
agricultural load is small relative to other nonpoint sources relative to
other nonpoint sources, controlling agricultural NPS pollutants may not
48
-------
Figure 6. Water uses, water quality indicators, and pollutants involved
in nonpoint pollution.
IMPAIRED WATER USES
Primary Contact Recreation
Secondary Contact Recreation
Wildlife and Aquatic Life
Drinking Water Supply
Agricultural Use
Industrial Use
WATER QUALITY INDICATORS
Human Sickness
Algal Blooms
Turbidity
Fish Kills
High BOD
Low Biological Diversity
POLLUTANTS
Sediment
Phosphorus
Nitrogen
Pesticides
Animal Wastes
Salts
49
-------
improve water quality significantly. Figure 7 shows the steps involved in
determining the controls necessary to meet water quality goals.
Since the determination of a control strategy is ultimately based upon
an analysis of cost-effectiveness, in the early stages of planning
rudimentary cost comparisons of controlling the various possible pollutant
sources should be developed. These comparisons will help determine which
sources should be treated first.
STEP 3: DETERMINING APPLICABLE CONTROL MEASURES
The process by which alternatives can be selected and evaluated, once
specific water quality problems have been identified, can be generalized for
all problem areas. Four basic screening criteria are involved:
1. The practices chosen must have the potential to control the
pathway of the problem pollutant . The primary pathway of nutrients strongly
adsorbed to soil and sediment is the soil erosion and sedimentation process,
whereas the pathway of moderately and weakly adsorbed pollutants is soil
erosion and overland flow. The primary pathway of nonadsorbed pollutants is
deep percolation and subsurface flow. A practice which acts to control one
or more pollutant pathways is termed a candidate measure. Candidate
measures can be ranked according to their ability to control the primary
pathway(s) of the pollutant(s) of concern. Figure 8 shows elements involved
in agricultural nonpoint source pollutant delivery; factors affecting
the three major pollutant pathways are indicated.
2, Candidate practices must be compatible with the farm management
system. Candidate measures have been classified in Section 3 as managerial,
vegetative, or structural controls. These categories reflect differences in
the permanence of practices; managerial and vegetative controls must be
renewed annually while structural controls generally involve a much longer
time horizon and some capital investment. The nature of water quality
management suggests the application and re-evaluation of progressively more
permanent measures. There are two principal reasons for this approach:
(a) It is likely in the early stages of water quality
management that specific nonpoint source contributions
will not be precisely established.
(b) Many water quality problems are reversible processes
providing the time for incremental practice application
and evaluation of resulting water quality improvements.
3. Candidate practices must be technically and economically feasi-
bile in the area of study. In most instances management controls will be
appropriate. Certain structural and vegetative measures may not be suitable
because of topography, soil characteristics, climate, equipment limitations,
or marketing factors (see Table 4). For example, contour farming (NIA-11)
is a widely used and effective soil erosion measure, but excessive slope
lengths limit its effectiveness. Contouring, in addition, is difficult to
implement on complex slopes or with certain large modern farm machinery. If
50
-------
Figure 7. Steps involved in determining the levels and types of controls
necessary to meet water quality goals.
Water Quality Goals
i
Pollution Problems Defined
Water Quality Standards I *f*" 1 Desired Beneficial Water Uses;
Load Reduction Needed to Meet
Water Quality Standards
f
Load Estimates
•
I
Point Sources ^"^ Non-point Sources
i Control Costs i—•»! Non-agricultural I I Background Load
T T
Treatment Costs
^™" Agricultural Load
Agricultural I
Base Load ^
Manageable Agricultural
Load
f Control Costs "" *l
i
Treatment Costs
51
-------
TABLE 4. DEGREE TO WHICH VARIOUS PHYSICAL VARIABLES LIMIT THE USE OF
CANDIDATE MEASURES
CANDIDATE MEASURES
Soil and Topography
0)
ft
o
r-t
CO
r^I
| 1
60
a
0)
a
ft
o
rH
CO
en
0)
ft
o
CO
X
cu
r-H
ft
e
o
u
en
rH
•H
o
to
*j
o
rH
,— (
cd
rC
CO
en
rH
•H
0
CO
13
0)
a
•H
cd
S-i
Q
K^
t-H
J-i
0
O
PX,
r^
O
•H
4J
O
cd
ft
B
o
u
rH
•H
O
CO
Equipment
4-1
a
0)
ft
•H
3
cr
W
, — 1
3
0)
a)
03
'Ct
,—J
Q)
•H
p&4
rH
|— {
flj
s
CO
}H|
O
0)
rH
o
cd
4J
•H
3
CO
4J
O
2
u
o
K^
o
(3
OJ
-iH
O
•H 4-1
IM C
MH OJ
w s
ft
4J -H
ti 3
0) cr
a w
ft
•H &
3 a)
cr ss
W
tn
03 (1)
a) J-i
0 -H
3 3
ij cr
CD C1J
Managerial Controls
Improved method of application
NIA-3
Improved timing of field tillage
operations NIA-4
Optimizing time of planting NIA-7
Using mechanical weed control
methods NIA-8
Reduced tillage systems NIA-9
Contour farming NIA-11
Vegetative Controls
Sod based rotations NIA-13
Contour strip cropping NIA-15
Structural Controls
Terraces NIA-19
Diversions and interception drains
NIA-20
Grassed Waterways NIA-21
+
0
0
+
0
*
0
*
0
0
0
0
0
0
0
0
*
0
*
0
0
0
+
0
0
*
0
*
0
*
*
+
+
0
0
0
0
0
0
0
0
*
0
0
0
*
+
0
*
*
0
0
0
0
0
0
*
0
+
+
0
0
0
0
0
0
0
0
0
0
0
+
0
*
*/+
+
+
0
0
0
0
0
*
0
*
0
0
0
0
0
0
0
0
*
0
*
+
+
0
*
0
0
+
*
0
+
+
0
0
0
0 = Not limited; + = Minor limitation; * = Major limitation
52
-------
Figure 8. Elements of the agricultural NFS pollutant delivery process.
FACTORS
AFFECTING
AVAILABILITY
OF POLLUTANT
AT FIELD SITE
POLLUTANT
CATEGORIES
POLLUTANT
PATHWAYS
Land Use
1 Material Inputs i
, Management ,
• Practices '
i , j
| % ORGAN
i MATTER
Strongly
Adsorbed
1C ' _ ~n.. •*
__— L SOIL pH
i »-SOIL-« — ____ r~~ey~~rfXv~
i 1 % CLAY
Moderately 1
Adsorbed |
Weakly
Adsorbed
Non-
Adsorbed
SOIL EROSION
AND
SEDIMENTATION
OVERLAND
FLOW >
FACTORS
AFFECTING
TRANSPORT
IN SOIL AND
WATER
Soil Erodibility
Slope and Slope Length
Rainfall Intensity and
Distribution
Vegetative Cover
Irrigation
Topography f- -
Distance to Water Body I— *•
LEACHING AND
SUBSURFACE
FLOW
Rainfall Intensity and
Duration
Soil Moisture
Infiltration Rate
Infiltration Capacity
Vegetative Cover
Slope
•*—"1 Enrichment
•« 1 Concentration in Water
| WATER BODY |
poorly drained fields are contoured, surface drainage may be even further
decreased. Certain practices, such as terraces (NIA-19) may not be
well-suited for modern farm equipment or areas where shallow topsoil
overlays less permeable subsoils. Although terrace design may be modified
to overcome these limitations in some cases, additional costs and risks
should be noted.
4. Candidate measures must be screened according to their fundamental
socio-economic feasibility. It should be emphasized that the effectiveness
of most NPS controls and the economic impact of practices are strongly in-
fluenced by crop rotations. The fact that certain crop rotations exist in a
53
-------
particular area often reflect both long term and short term market adjust-
ments by farm operators. Changing rotations or introducing new crops as a
control measure such that farm crop production is decreased should, in most
cases, be discouraged. Where these changes are technically necessary or
more profitable than a structural alternative, a comprehensive study of the
marketing implications should be undertaken.
Using the four criteria discussed above, a list of candidate measures
can be assembled for a given watershed or sub-watershed. This list is the
end product of Step 4. In Step 5 below, this list is screened further.
STEP 4: CHOOSING THE UNIT OF ANALYSIS
The watershed is a logical unit of analysis where pollutant load esti-
mates and water quality data can be related. Monitoring problem pollutants
and evaluating candidate practices is very much simplified if the analytical
unit is a watershed rather than a political boundary. However, none of the
analytical methods described in this section or the case studies in Section
5 require application to the total watershed. All of the computational
methods used simply apply to a defined land area, not necessarily draining
to the same stream.
A watershed can be divided Into subwatershed areas. This allows com-
parisons to be made between subwatershed areas concerning the need for and
effect of Implementing nonpoint source control measures. Program emphasis
can then be directed toward those areas where nonpoint source controls are
most efficient.
Although a watershed is an appropriate unit of analysis with respect to
the evaluation of nonpoint source controls, there are some important limita-
tions. First, soil groups, farms and even fields can overlap more than one
watershed. Second, much of the soil and agricultural statistics and data
needed to evaluate practices are not collected on a watershed basis. These
limitations are discussed in Appendix F; they have, in part, been accounted
for in the following analyses.
In evaluating NPS controls at the watershed level it is generally
assumed that controls do not affect the prices paid or received by farmers.
This assumption is not valid for large areas (eg., corn belt) where widely
adopted NPS controls would significantly reduce the market's supply of com-
modities or farm inputs.
STEP 5: ESTABLISHING THE BASE CONDITION
The choice of efficient nonpoint source controls requires that the
cost-effectiveness of candidate practices or sets of candidate practices be
compared. For the methods presented in this manual, cost-effectiveness is
defined as the change In farm income compared to the change in delivered
pollutant load caused by practice Implementation (Figure 9).
54
-------
Figure 9. Steps involved in determining cost-effectiveness.
BASE CONDITION
Estimate Pollutant
Delivery
Estimate Net
Income
POST-CONTROL CONDITION
Estimate Pollutant
Delivery
Estimate Net
Income
Calculate Difference
Between Two Delivery Estimates
(Change in Delivery)
Calculate Difference
Between Two Net Income Estimates
(Change hi Net Income)
\ /
Calculate
Change in Net Income
Change in Pollutant Delivery
To determine the cost-effectiveness of a particular candidate measure,
the pollutant loading and farm income before and after practice implementa-
tion must be determined. The present farm management system and the assoc-
iated pollutant load should serve as the basis for the comparison of alter-
native systems. For example, if fall plowing is the common practice in the
watershed, it should be considered as one of the base management practices.
Since costs and impacts of applying most BMPs will vary widely depending on
the productivity, slope and drainage of soil, cost effectiveness must be
determined for each important soil or for groupings of similar soils.
Calculating Pollutant Loading
To determine pollutant loading from a watershed, overland and subsur-
face water flows and the soil loss for each soil group in the watershed must
be approximated. Conceptually, the movement of surface and subsurface water
from cropland to streams can be described as follows:
Uu + Ou = Pu + Iu - Eu + Bu + Du
Where: Uu = subsurface flow (cm) during time interval t
Ou = overland flow (cm)
Pu = precipitation (cm)
Iu = irrigation application (cm)
Eu = evapotranspiration (cm)
Bu = change in surface storage (cm)
Du = change in subsurface storage (cm)
55
-------
The amount of soil loss can be estimated using the universal soil loss
equation (USLE) or some modification of USLE:
Au = RK(LS)CP
Where: Au = soil erosion (kg/ha) during time interval t
R = rainfall erosivity factor
K = soil credibility
(LS) = slope length factor
P = conservation practice factor
A number of factors strongly influence the distribution and quantity of
the soil and water losses. Soil drainage and hydrologic condition, precipi-
tation duration and intensity, soil credibility, slope and slope length,
field capacity, crop and cropping practices and conservation practices are
examples of these factors. After water flows and soil loss have been esti-
mated, pollutant loading calculations can be made by assuming a level of
pollutant availability, of mixing with the soil and water components, and of
transport to a water course. Appendices A, B and C include a more detailed
discussion of techniques which can be used to estimate soil, water, and
related pollutant losses.
As discussed in Section 2, pollutant availability and transport involve
complex processes. For example, inorganic nitrogen (N) is made available
from soil each year and the decay or mineralization of manurial organic
matter of crop residues add to the pool of inorganic N. In addition, the
actual concentration of inorganic N in a particular soil may change after
precipitation events, as transformations take place between different forms
of N and as volatilization and denitrification losses occur. A set of
particular crop, soil, and climatic conditions determine plowing, planting,
and emergence dates as well as fertilizer needs, crop uptake, and yield.
These conditions determine fallow periods, rate of canopy development,
amount of surface residue and, consequently, the availability of nutrients,
pesticides, and salts for transport.
Pollutant Delivery --
Pollutant delivery to streams can be defined as the fraction of a pol-
lutant leaving a defined area (e.g., field) which actually reaches a certain
point in the stream in the time interval under consideration. In the case
of sediment, the delivery ration is given as the quantity of suspended
solids at the mouth of the watershed. It is a function primarily of the
topographic and vegetative character of the terrain. Where opportunities
for redeposition are greatest, the pollutant delivery ratios are lowest.
In applying the concept of pollutant delivery to substances being
transported with overland or subsurface flows, the receiving stream must
first be defined. A receiving watercourse can be defined as any channel
where concentrated flow is evident, even if intermittent. This is a broader
definition than merely a stream where water quality is monitored or where
the quality of water has a direct impact on designated water use. Figure
10 illustrates examples of points that might be designated as receiving
56
-------
watercourses. Point A, at the mouth of the watershed, represents the cumul
ative contribution of the entire drainage area from both overland and sub-
surface flow. Point B designates a point where a roadside drainage ditch
contributes intermittently to main stream flow. Point C is the location
where overland flow from Field 1 drains into a roadside ditch. Point D
represents direct field discharge, and Point E represents discharge across
an intervening land surface. Overland flow leaving Field 2 near Point D,
has a greater opportunity for infiltration, ponding and redeposition of
suspended solids than does flow leaving Field 3 near Point E.
INTERMITTENT
''STREAM
BRANCH
TRIBUTARY
MONITORING
STATION
ROAD
MAIN
TRIBUTARY
Figure 10. Examples of alternative points for the designation of a
receiving water course.
57
-------
In the case of subsurface flow, the definition of a receiving stream or
watercourse becomes more ambiguous. Although much of the water infiltrating
soil may eventually reappear as surface flow, the pathway and time frame for
this process can vary greatly. Infiltrating water can reappear as interflow
(Pathway A of Figure 11) or can percolate to the underground water table and
flow to a stream (Pathway B). Some subsurface flow may percolate to
aquifers, as Pathway C in Figure 11, but only reach a water course after
considerable time. The final form and quantity of highly soluble, conserva-
tive substances like chlorides are not significantly affected by the partic-
ular pathway taken. For substances that are either weakly retained by the
soil or that degrade over time, however, the length and transit time of dif-
ferent pathways may affect actual pollutant delivery markedly.
Figure 11. Pathways of subsurface flow.
58
-------
Estimating the Cost of Practices
Perspective of Cost Estimates --
Estimates of the cost of nonpoint source pollution controls differ
according to the perspective taken. Control costs should Include instal-
lation, operation and maintenance expenditures. From the farm operator's
perspective, control costs should also include reductions in farm income
resulting from farm investment in water pollution controls. From society's
viewpoint, increased prices of agricultural products resulting from nonpoint
source controls, as well as costs of administering programs by public agen-
cies and changes in tax revenues, represent costs of improving water qual-
ity. In all cases there are opportunity costs associated with alternative
investment options. For example, water quality may be improved more per
unit of expenditure by investing in either point or non-agricultural non-
point source controls. Likewise, a farm operator's income may be reduced
more with certain practices than with others, or the social cost of in-
creased prices of agricultural commodities may be less desirable than the
increased product prices associated with point source controls.
The level of analysis used for this manual does not incorporate
national, regional or other social cost perspectives. It is assumed that
prices paid to farmers for agricultural commodities and paid by farmers for
goods and services do not change with implementation of nonpoint source
pollution controls. All cost estimates generated with this methodology
include direct installation and maintenance costs, as well as estimated
changes in farm income. By estimating farm income effects, practices which
are more likely to be accepted by farmers can be selected. Cost estimation
is discussed in more detail in Appendix F.
Base Income Level --
A base level of farm income must be established so that the relative
cost of candidate practices can be compared. Because the implementation of
nonpoint source control measures does not affect all aspects of a farm
enterprise, only certain returns and expenditures must be budgeted. This
technique of partial budgeting can save both time and planning resources.
Appendix F describes these budgeting procedures in detail.
The fixed and variable costs of production which are most affected by
practice implementation include machinery, irrigation, labor, pesticide, and
fertilizer costs. These costs are quantified for the base condition for
each area studied since they vary from farm to farm as management practices
and machinery compliments vary. Since each crop has different management
requirements, cost estimates are conveniently done on a crop basis.
The base farm management system should be set up to achieve the maximum
level of net income possible given the site conditions and current techno-
logy. The base condition generally approximates profit maximizing farm
operations. Losses of pollutants are estimated for this level of farm
income and farm activity in the watershed. The relative changes that occur
from this base condition are then used to evaluate the cost-effectiveness of
control practices.
59
-------
STEP 6: EVALUATING CANDIDATE CONTROL MEASURES
Qualitative evaluations do not enable one to compare and choose NPS
control measures which are best suited for a particular farming enterprise
or set of field conditions. Some quantitative estimates of the effective-
ness and costs of the measures, therefore, are needed. Analytical proce-
dures to predict sediment, nutrient and salt losses with implementation of
selected control practices have been developed and are described in this
manual. It is difficult, however, to quantify pesticide losses because of
inherent characteristics of such chemicals. The selection of candidate
measures for the control of pesticide losses has therefore been approached
qualitatively. Persistence, toxicity, drift, and volatilization vary for
different pesticides and for varying soil, water, and wind conditions. In
many cases, the quantity, destination and toxic effects of pesticide volati-
lization and drift are unknown. Quantitive methods which may be used to
estimate pollutant loads include the Universal Soil Loss Equation (USLE) for
sediment losses, SCS Curve Number for runoff leaching losses, and simulation
modelling for nutrient losses. Partial budgeting and linear programming are
used to estimate net income. These methods are discussed below.
Sediment Loss From Non-Irrigated Fields
Sediment loads to a stream can be estimated by calculating gross field
erosion and an appropriate sediment delivery ratio (SDR). A common method
for estimating rill and inter-rill erosion is the Universal Soil Loss Equa-
tion (USLE). The USLE predicts, for a specific area, the average annual
rill and inter-rill soil erosion created by rainfall and associated runoff.
The land area being evaluated can consist of either a single field or a
larger area. The predictive accuracy of the USLE, however, decreases for
slope lengths greater than 122 meters. Details of the USLE and how it can
be used to predict erosion and adsorbed pollutant losses are given in Tech-
nical Report A.
Sediment Loss from Irrigated Areas
Although irrigation practices are designed to result in greater control
over water use and movement, soil erosion and sedimentation are evident on
all furrow-irrigated lands. Sprinkler irrigation systems can result in
localized runoff and erosion problems, such as when heavy application occurs
near the periphery of center-pivot sprinkler systems of moderate to fine-
textured soils. However, such runoff generally accumulates in the same
field or in nearby depressions.
Applications of the USLE in irrigated areas are limited. One approach
for estimating sediment loss was developed by Gossett (1975). A base-level
combination of crop, irrigation, and soil conditions is assumed, for which
sediment loss has been measured. Other crop, irrigation, and soil condi-
tions are then treated as adjustments, in the form of multipliers, to the
base-level estimate. This approach was used to predict erosion for the
60
-------
Yakima case study in Section 5. Computational methods for estimating sedi-
ment loss from irrigated cropland can be found in Appendix B.
Estimation of Runoff and Leaching Losses
Runoff from non-irrigated cropland can be determined using a number of
computational methods. One method is the curve number approach. Soil
moisture is computed by quantifying all significant inputs and outputs of
water. When a precipitation event occurs, water infilitrates into the soil
until soil water capacity is exceeded. The remaining rainfall is runoff or
is temporarily stored in surface depressions. The partitioning of water
between surface and subsurface flows is determined using the curve number,
an index reflecting the potential for runoff as determined by soil hydro-
logic group, crop, soil management practices and antecedent moisture condi-
tion. This method is described in Appendix A.
Runoff from irrigated cropland can be predicted with greater accuracy
than non-irrigated cropland. Water application, evapotranspiration and sur-
face return flow can be estimated directly. The quantity of water which
percolates is usually assumed to be the residual loss. Runoff volume can be
determined in a way similar, to sediment loss in irrigated agriculture. That
is, a base level of runoff is assumed and multipliers are used to account
for stream size, length of set, soil type and field slope. Runoff estimates
for these different conditions can be found in Appendix B.
Nutrient and Salt Losses from Non-Irrigated and Irrigated Croplands
Once soil and water movement have been calcultated, nutrient and salt
loading estimates can be made. These calculations account for the availabi-
lity of different forms of nitrogen (N) and phosphorus (P) in the soil pro-
file, equilibrium concentrations of N and P either associated with the soil
or in solution, and changes in concentration during storm events. Loading
estimates are typically made by multiplying average or flow dependent con-
centrations by the calculated runoff or leaching volume. Detailed examples
of nutrient loss estimates are made in Section 5 and further details are
given in Appendices A, B and C.
Economic Evaluation of a Farm Enterprise
Farm operations are generally made up of a number of enterprises.
Cropping enterprises may include the production of corn silage, corn grain
or winter wheat. Livestock enterprises may include milk, beef or swine
production. Each of these enterprises has a particular cost structure.
That is, all crops have variable growing expenses (seed, fertilizer, equip-
ment), whereas building use, land cost, insurance and interest are fixed for
a given farm and do not vary appreciably among crops or for a particular
year. A livestock enterprise will have a different set of variable and fix-
ed costs. Bedding, breeding and veterinary charges are examples of variable
costs, and manure disposal and feed storage are examples of relatively fixed
expenditures for a typical livestock operation.
61
-------
Implementation of nonpoint source pollutant controls changes the cost
structure of farm enterprises. It may also reduce total production. Calcu-
lation of farm Income changes thus involves an estimation of changes in cost
structure and total production.
Changes in management and vegetative practices require little or no
capital in the form of construction or equipment. Since all management
practices apply only to existing operations, minimal direct costs of imple-
mentation, operation, or maintenance should be required. Establishing
vegetative cover usually requires only tillage, growing, and harvesting
expenditures. These expenses may be considered to be indirect costs of the
practice if they are associated with sod or grain grown in rotations as part
of existing crop enterprises. Structural practices, on the other hand,
usually necessitate earthmoving and/or construction activities, which
involve installation, operation, and maintenance costs.
Programming Method
Whether total or partial farm budgeting methods are used to estimate
the cost of nonpoint source pollutant controls, the many enterprise and
practice combinations can present a bewildering array of options. Compu-
tational assistance is now available through a number of standardized pro-
gramming techniques. One of the most common methods used is linear pro-
gramming (LP). Linear Programming methods are particularly appropriate for
comparing agricultural investment decisions, since numerous combinations of
crops, tillage methods, fertilizer management and other activities may be
evaluated for each field. This technique is discussed in detail in Appendix
F as well as illustrated in the examples of Section 5.
STEP 7: DEVELOPING AN OPTIMAL CONTROL STRATEGY
Calculating cost-effectiveness for each individual practice and combin-
ation of practices for each subwatershed unit is generally not feasible.
Time and analytical resources usually limit the number of options that can
be evaluated. In order to insure the evaluation of a variety of realistic
control options, sets of practices can be assembled from those screened in
Step 3. Subjective judgment concerning the compatability of certain prac-
tices to farming conditions in the study watershed must be used to reduce
the number of alternatives considered. It is important to encourage the
consideration of a range of practices. The grouping of practices in Table
2 into management, vegetative and structural controls illustrates the range
that might be considered.
The cost-effectiveness evaluation of these sets of practices for dif-
ferent subwatersheds is the basis for developing an "optimal" NPS control
strategy. There are a number of techniques which can be used to compare
practice sets. Three techniques which can be used are: 1) the marginal
adjustment cost method, 2) additive optimization, and 3) the efficiency
frontier method. They are discussed below and are demonstrated in the case
studies presented in Section 5 and in the accompanying Technical Reports.
62
-------
Marginal Adjustment Cost Method
For a particular load reduction or level of pollutant load, the mar-
ginal cost to change or adjust farm activities can be estimated. The
marginal cost (MC) is defined as the additional cost incurred by reducing
pollutant load one more unit. For example, if the total cost of load reduc-
tion is $500 for x units but $510 for x + 1 units, the marginal cost at x +
1 units is $10. Figure 12 illustrates this graphically. Point A corres-
ponds to a pollutant load reduction of B and a change in farm income of C.
The slope of the curve at Point A is the marginal cost of control. This
particular curve shows a constantly increasing MC; as pollutant loading is
restricted to lower and lower levels, cost increases at an increasing rate.
The average unit cost (OC/OB), defined as total change in farm income divid-
ed by total change in pollutant load, thus increases as pollutant loading is
reduced.
Figure 12. Marginal and average adjustment cost of pollutant
load adjustments.
o
I-
o
o>
>
o
0.
o
w
O
Jt
o
o
CD
POLLUTANT LOAD REDUCTION, APOLL
63
-------
Marginal adjustment costs can be expected to vary for different water-
sheds, subwatersheds and sequences of practice treatment. Consider Figure
13. Marginal cost curves have been generated (after screening practices to
determine those appropriate for the watershed) for three sets of practices:
managerial (M), vegetative (V) and structural (S). In order to optimize the
allocation of water pollution control funds, each unit of pollutant
reduction should be achieved at least cost. This is accomplished by
treating first those subwatersheds with the lowest marginal costs. In Fig-
ure 13 these would be managerial controls in subwatersheds 1 and 4.
Figure 13. Marginal cost curves for sets of practices on a
subwatershed basis.
S: Structural
V: Vegetative
M: Management
64
-------
The cost-effectiveness of practices also depends on the desired level
of control. Certain practices have large initial implementation costs, but
the unit control costs decrease at higher levels of pollutant reductions.-
If a relatively large pollutant load reduction is needed, these practices
can be more cost-effective than practices with a low initial cost since
the average cost may be lower.
The effectiveness of practices in an area may be influenced by other
factors such as precipitation. For example, in Figure 14, practice "A"
(contouring) may be highly effective during rainfalls of low intensity
whereas practice "B" (terracing) is not as cost-effective. However, a more
intense precipitation event may show practice "B" more effective than "A"
(other things being equal).
Figure 14. Relationship between cost-effectiveness and rainfall
intensity for two different practices.
o
o.
CO
CO
LO
z
UJ
o
ID
U_
U.
LU
CO
O
Practice A
Practice B
RAINFALL INTENSITY
Additive Optimization Method
The approach of determining optimal control investments is best exem-
plified with respect to salt control in irrigated areas. The approach is
demonstrated in the Grand Valley case study of Appendix C.
65
-------
The steps involved in the additive optimization process are given in
Figure "15. Alternative 1 is a particular set of practices applied to a
given subwatershed area. Alternative strategy 2 is a set of practices
applied to another area. Alternative 1 might involve canal levelling and
lining measures while alternative 2 might involve the application of on-farm
irrigation scheduling and changes in irrigation systems for a particular
irrigation district in the watershed. Level 1 practices are specific
practices, while level 2 and 3 are combinations of practices. The cost of
implementing alternatives at various levels of effectiveness are estimated.
The minimal cost of aggregate reductions in pollutant load can then be
determined as a combination of level 1 practices. Subsequent load
reductions for levels of practice implementation are determined in the same
manner (levels 2, 3, and 4). This procedure assumes independence between
practices and subwatershed units; it is assumed that the application of one
practice alternative in a given area does not change the cost or
effectiveness of other practice applications.
Efficiency Frontier Method
The efficiency frontier technique involves plotting the minimal cost of
control associated with different levels of pollutant load reduction. An
efficiency frontier is a locus of points drawn from the evaluation of alter-
native NPS controls. The curve defines least costly practices at different
control levels. A rough approximation can then be made of total pathway
control costs for a given level of load reduction. The Yakima Valley case
study in Section 5 demonstrates the use of this technique. Figure 16 shows
examples of cost-effectiveness functions. Points A-E are alternative
control strategies; A might be contouring and strip cropping and B no-
tillage and reduced tillage on selected soils. The efficiency frontiers can
be used to evaluate each control strategy with respect to effectiveness in
reducing soil loss, overland flow, nitrogen loss, and phosphorus loss
simultaneously. Efficiency frontiers result in an estimated cost for each
control level. Curve BE in graph 1 of Figure 16 represents the most
cost-effective practices considered. Whereas practice E is relatively
distant from the efficiency frontiers for controlling overland flow and
nitrogen loss, practice B is relatively effective at all phases of control.
Thus, for the pollutants and practices considered, practice B is the most
efficient control strategy for complete pathway control. The simulation
modeling and linear programming methods presented in the case study
watersheds of Section 5 appear particularly well-suited to this evaluation
methpd.
Although the proceeding sequence of data collection and evaluation
methods are suitable for all water pollutants, pesticides have peculiar
characteristics and therefore are treated separately. Computational methods
for the estimation of pesticide losses exist, but are generally not as well-
established and to a large extent remain unvalidated. The reader is
referred to Davidson (1975) and Bailey (1974) for examples of pesticide
evaluation methods.
66
-------
Figure 15. Steps in the additive optimization process.
Desired Level
of Salinity ,
Control ;
ALTERNATIVE I ,level 3
COST-EFFECTIVENESS
FUNCTION
LOAD REDUCTION
.'
ALTERNATIVE I , ^vel 1
alt.l.level 2 investments
lt.2, level 2 investments
/
Level I
--Costs
LOAD REDUCTION (A Poll)
Optimal Level 2
Cost from level 3
ALTERNATIVE 2, level 2
LOAD REDUCTION (A Pol I)
Optimal level 2
Cost from level 3
* LEVEL 1 COSTS ARE SPECIFIC PRACTICE COSTS
67
-------
Figure 16. Efficiency frontiers showing costs of five control
strategies for four different pollutants.
A SOIL LOSS
AOVERLAND FLOW
A PHOSPHORUS LOSS A NITROGEN LOSS
EVALUATION OF PESTICIDE CONTROLS
Agricultural use of pesticides includes control of plant disease,
insects, mites, nematodes and weeds which damage crops. Like crop nutri-
ents, residual amounts of pesticide can be transported from treated fields
to receiving streams in overland or subsurface flow. Similarly, as in the
case of nutrients, a precise relationship between agricultural use of pesti-
cides and subsequent changes in water quality is difficult to establish. It
is particularly difficult since thousands of registered pesticides are
used. Also, in general, relatively small quantities are applied to cropland
making detection and monitoring impractical. Toxicity, persistence, and
other chemical properties, in addition to soil adsorption, and solubility
interact to influence a pesticide's impact on water quality (Figure 17).
The influence which these properties have on
transport is briefly described below. Appendix E
Losses) provides additional details.
Grouping Pesticides
pesticide availability and
(Control of Pesticide
In order to begin considering the control of pesticide losses from
cropland, it is necessary to systematically characterize the many chemical
compounds which are used to control crop pests. Table 5 lists common
chemicals used to control major target pests. Although general statements
68
-------
can be made regarding the behavior of pesticides from a specific group, the
characteristics of an individual pesticide are unique and properties of each
group will not necessarily apply to every pesticide in that group. In addi-
tion, some controls will be limited in their use. For these reasons, the
individual properties of each specific pesticide to be controlled should be
evaluated, prior to initiating a management program, in order to insure that
the management practice that is selected is best suited to control the
existing or potential problems.
Figure 17. Factors affecting the transport and water Quality
impact of a pesticide.
toxicity
persistence
soil adsorption
solubility
other chemical
properties
drift
PESTICIDE
GROUND WATER
(receiving water)
LAKE
(receiving water)
Persistence
One important characteristic of a pesticide is its persistence. One
definition of persistence is that period of time necessary for the complete
degradation of the material into harmless products. Since the primary mode
of degradation is generally biochemical, factors which reduce biological
activity decrease the rate of degradation. These factors include low soil
moisture, oxygen content, temperature and organic matter content, and ex-
treme pH (Figure 18). The relative influence of each factor depends on the
specific pesticide and the site conditions.
69
-------
[DEGRADATION PROCESS^>
PH
Matter
i i
i Temperature i
I i
i J
| Oxygen J
i Content i
I J
Figure 18. Factors affecting pesticide persistence.
Strongly adsorbed pesticides are relatively immobile in the soil pro-
file. The attenuation of pesticides by soil particles allows for degrada-
tion and dissipation of its toxicity. Strong adsorption of pesticides to
soil particles will generally keep pesticides close to the surface where
they were applied. The soil surface is the area of greatest biological ac-
tivity and therefore pesticide degradation rates are likely to be greater.
Organochlorines are, in general, some of the most persistent pesti-
cides. Organophosphates, carbamates and most herbicides are more easily
degraded. The triazine herbicides are degraded mainly by chemical action
but can persist for substantially longer periods of time. Table 6 lists the
half-lives of some common pesticides. Half-life is defined as the amount of
time required for 50% of the original material to disappear. For
approximately 94% of the material to degrade, it would take four times as
long. Half-lives vary under different biological, chemical and physical
conditions. For example, a microorganism population may adapt to the pres-
ence of a certain pesticide, thus causing an increase in the degradation
rate and a decrease in the pesticide half-life. This phenomenon of adapta-
tion can increase the need for multiple applications.
70
-------
TABLE 5. SELECTED GROUPS OF INSECTICIDES. MITICIDES AND FUNGICIDES
INSECTICIDES AND MITICIDES
Organochlorines
Chemical Compounds
Trade Name
Organophosphates
Carbamates
Farmamidines
Organotins
FUNGICIDES
Inorganic
Organic
DDT
Cyclodienes
Polychloroterpenes
Phosphonodithioates
Phosphonothioates
Phosphorothioates
Phosphorodithioates
Coppers
Dithiocarbamates
Phthalimides
Other
Methoxychlor
Chiorobenzilate
Eudosulfan
Chlordane
Aldrin, Dieldrin
Toxaphene
Fonofos, Terbufos
EPN
Parathion, Methyl
Parathion, Diazinon
Dursban, Demeton
Ma lathion,
Azinphosmethyl
Dimethoate, Phofate
Disulfoton
Methomy1
Carbofuran
Carburyl
Aldicarb
Chlorodimeform
Cyhexatin
Copper sulfate
Maneb, Ferbam
Captan, Difolatan
Benomyl, Chlorothalonil
71
-------
TABLE 6. HALF-LIVES OF SOME COMMONLY USED PESTICIDES
Pesticide Approximate Half-Life (Weeks)
Lead, arsenic, copper, mercury 500 - 1500
Dieldrin 100 - 200
Triazine herbicides • 50 - 100
Benzoic acid herbicides 10 - 50
Urea herbicides 15 - 40
2,4-d; 2,4,5-t, herbicides 5 - 20
Organophosphate insecticides 1 - 10
Carbamate insecticides 1 - 5
Toxicity
Pesticides often are toxic to non-target organisms as well as target
organisms. The quantitites of pesticides reaching the nontarget area, the
likelihood (or risk) of exposure of organisms, and the sensitivity of the
organism coming into contact with the pesticide must be considered. The
impact on specific organisms must be seen within the context of their posi-
tion within a food web. Although a pesticide may be only moderately toxic
to a fish species, high toxicity to organisms serving as fish foods may have
an equally detrimental effect on the fish populations. The dose-response
impact of a pesticide must also be seen within the context of environmental
conditions normally encountered by the test organism. Temperature extremes,
nutrient levels and unusual pH may all render an individual pesticide more
or less toxic to the test organism. Such synergistic toxic effects are
possibly the rule rather than the exception in nature. These effects, how-
ever, are poorly understood and extremely difficult to define since the
possible combinations of environmental conditions are practically infinite.
Solubility, Adsorption and Transport Characteristics
Solubility is also an important characteristic of a pesticide. Certain
pesticides are fat soluble while others are water soluble. Most organophos-
phates, for example, have low fat solubility. Pesticides which are fat
soluble, such as some organochlorines, become concentrated in the fatty
tissues of organisms, often in higher concentrations than present in the
surrounding environment. Animals which are higher in the food chain are
more likely to accumulate pesticides because fat soluble compounds tend to
concentrate at higher levels in a food chain. This process of biomagnifica-
tion can cause severe problems of survival for a species.
72
-------
Pesticides move from agricultural fields or dumping areas to streams
and lakes by drift, volatilization, in the surface flow (suspended or dis-
solved), attached to eroding sediment, and dissolved in subsurface flow.
Accidental spills or dumping of excess chemicals into surface waters also
contribute substantially to pesticide levels in the water.
The quantity of pesticide lost by drift is dependent on the method of
application; small liquid drop or dust applications increase losses whereas
granular applications tend to decrease losses. Climatic conditions at the
time of application are also important; high winds tend to disperse pesti-
cide molecules rapidly. Volatilization depends on certain physical and
chemical properties of the pesticide (i.e., vapor pressure, water solubil-
ity, adsorption coefficient), and on the moisture content and temperature of
the soil. Volatile pesticides are frequently regarded as more desirable
since they infiltrate soil pores more eaasily, and thus more effectively
reach the target pest. However, from the viewpoint of total environmental
quality, volatile pesticides may present a greater environmental risk.
Volatilization increases pesticide loading to the atmosphere and the possi-
bility of eventual redeposition to streams and lakes. In addition, the loss
of pesticides by the volatilization process can be considered an economic
loss since the farmer is paying for chemicals which do not end up in the
target area. It has been estimated that half of the pesticides applied to
field crops enter the atmosphere through vaporization from plant and soil
surfaces.
Soil incorporation, cover crop, high clay and organic matter content,
and deep plant penetration reduce volatilization. However, the rate of
degradation of a pesticide may be decreased (the persistence increased) if
the applied chemical is incorporated into the soil.
Surface and subsurface losses are determined by the adsorption and
solubility properties of the pesticides, the characteristics of the site,
and the timing, mode, placement, and rate of chemical application. Many of
these factors can be manipulated to reduce the impact that the agricultural
usage of pesticides has on water quality.
Pesticide mobility is determined in part by various factors which
encourage soil adsorption. Soil adsorption tends to increase under higher
soil moisture content and lower ph. In general, herbicides tend to have
lower adsorption coefficients than insecticides or fungicides, although
pesticide solubility varies considerably within groups.
The adsorption characteristics of the pesticide determine the mode of
transport. Soluble and weakly adsorbed pesticides usually percolate down-
ward through the soil profile. Although this is the most common pathway for
these pesticides, intense rainfall, steep topography or saturated soil con-
ditions may cause them to be lost via runoff flow. Moderately adsorbed
pesticides are primarily transported in surface flow but are also lost with
eroded sediment. Very strongly adsorbed pesticides like trifluralin and
toxaphene are transported primarily on eroded sediment. Appendix E
describes adsorption characteristics of a number of commonly used insecti-
cides.
73
-------
The method of application affects pesticide transport processes and
thus pesticide losses. Applying a moderately adsorbed pesticide on the soil
surface as opposed to below the surface increases the potential for loss.
In contrast, soluble pesticides are more likely to be leached if they are
incorporated into the soil surface. The timing of an application is also
important; chemicals applied shortly before precipitation events have a
greater potential for loss.
FACTORS INFLUENCING SUCCESSFUL IMPLEMENTATION OF NPS CONTROL PRACTICES
Social Costs and Benefits
The impetus for the implementation of measures to prevent or reduce
water pollution lies ultimately with the public. It is through public input
that legal measures such as effluent standards, effluent taxes, and
subsidies are instituted. These measures are the result of a cost-benefit
analysis, in which social costs associated with implementation and with any
decreases in production, and social benefits resulting from improved water
quality have been weighed. Appendix F discusses these trade-offs and
describes in greater detail alternative agricultural pollution control
policies.
Farm Profit Maximization
Cost effectiveness analyses have typically assumed that the farm
manager is maximizing short-run profits (MSRP). He adopts a particular
control practice or set of practices based upon the immediate opportunity
cost (i.e., the perceived value of how the money could be used elsewhere) of
abatement. That is, the MSRP approach assumes the decision to adopt and
implement various control practices is made on the basis of their short-run
profitability. At any level of abatement, other things being equal, the
efficient practice, from the farmer's standpoint is that which impacts
profit least.
The MSRP approach does oversimplify the producer's decision of adoption
or nonadoption by relying solely upon a static concept of production
efficiency. It is clear that not all farms adjust to and adopt new
technology or environmental controls equivalently. Such considerations as
financial structure of the farm, managerial objectives and capacity, farm
size, legal organization of the farm, tax treatment, and ability to bear
risk may figure prominently in the actual control practices adopted and
the speed of adoption.
The fact that a producer's net worth, disposable income, time horizon
and other single factors affecting decision-making are oversimplifications
has led to attempts to quantify investment decisions using what have been
called multi-variable methods. For example, Carter and Cocks (1975) devel-
oped a model which included both short term and long-term objectives. The
seven objectives included:
74
-------
1. Maximization of the present value of future consumption.
2. Maximization of the present value of future profits, where
profits are withdrawn each period.
3. Maximization of the present value of future profits, where
profits are reinvested each period.
4. Maximization of the discounted cash flow rate of return.
5. Maximization of the present value of future cash flow.
6. Maximization of terminal net worth.
7. Selection of the most preferred point on an efficient locus
showing present consumption versus terminal net worth.
Different results were found for all seven goals leading Carter and Cocks to
conclude that the choice of goals is critical in modeling farm behavior,
i.e. farmer decision-making. This they found to be particularly true of a
consumption net worth tradeoff goal.
Choosing the appropriate objective or criteria will influence the
cost-effectiveness ranking of practices. For a single watershed, the size,
financial structure, and investment options may be reasonably similar for
all farms. However, when this is not the case, the variation in decision
criteria should be recognized if implementation efforts are to be success-
ful. The economic implications of choosing a particular criteria are dis-
cussed in detail in Appendix F.
Farm Characteristics
Farm investment decisions are often Influenced by tax laws. These laws
bias investment decisions depending on the objectives of the farm operator.
Three common strategies taken by farm operators include:
1. Reduction of actual out-of-pocket costs,
2. Deferment of income taxes,
3. Conversion of ordinary income to capital gain.
Farm size can also be an important factor influencing tax strategy and
the types of NPS controls farmers are likely to invest in. Capital inten-
sive controls may be more appealing to large farms with tight labor con-
straints while labor intensive practices may be more appealing to small
farms. In addition, a weak equity position may discourage both large and
small farms from investing in capital intensive controls. Figure 19
illustrates factors affecting farmer adoption of BMP's.
Risk and Uncertainty
Individual producers will exhibit widely varying perceptions of the
risk associated with different control practices. Changes in technology,
legal and institutional structure of farm services, yield and price varia-
tions can all influence the profitability over time.
The extent to which tax strategy, farm size, and risk preference can be
incorporated into NPS control program are not clear. Appendix F discusses
these factors in detail.
75
-------
Figure 19. Factors affecting the adoption of a BMP.
Equity
Position
Tax
Laws
Labor
Intensive
DECISION
TO
ADOPT BMP
Cost-Sharing Incentives
One common method of encouraging the implementation of NPS controls is
through subsidy payments. This system has been used for some time in con-
trolling soil erosion. These payments usually share in the cost of practice
construction or installation with the farmer expected to bear the full cost
of maintenance.
Cost-sharing formulas can be designed to account for some of the fac-
tors which influence the successful implementation of NPS controls. For
example, cost-sharing arrangements directly affect decision criteria such
as short-run profits, net worth and debt/equity ratio. In addition, factors
such as farm size, tax strategy and risk/uncertainty preference can be
strongly influenced by different cost-sharing formulas. Despite opportuni-
ties for using cost-sharing as a means to account for different financial
preferences of farmers, little has been done in the past to incorporate
these preferences into subsidy payments for practices. Generally, cost-
sharing percentages have been uniform for all farms with strong bias toward
capital investments. In addition, cost-sharing formulas are always based on
installation and maintenance cost and not changes in income. An obvious
advantage with such cost-sharing rules is the small cost of administration
and practice monitoring. The principle disadvantage is that the cost-shar-
ing program may elicit adoption patterns not commensurate with the problems
to be solved.
76
-------
Perhaps the major limitations to current cost-sharing criteria are that
they generally exclude management practices. As demonstrated in Section V,
these types of practices are often the most cost-efficient NPS agricultural
controls. Therefore, limiting cost-sharing to structural controls can only
lead to an inefficient water quality program.
A complete discussion of the advantages and disadvantages of alterna-
tive cost-sharing programs can be found in Appendix F.
SUMMARY
The options available to water quality planners are numerous. The con-
trol practices described in Section 3 are general categories; many vari-
ations are possible. Cost-effectiveness, suitability, and likelihood of
practice acceptance are extremely important considerations in control
strategy selection. The nature of water quality mangement and the limita-
tions of data availability suggest an incremental program where practices
are implemented, their effects are monitored, and then practices are refined
over time. This approach is consistent with water quality goals for a num-
ber of reasons, two of which are:
1. Many water quality problems are reversible processes.
2. The complex interaction of physical and biological processes
in the transport of pollutants in space and time limit our
knowledge of the cause-effect relationship between practice
implementation and changes in water quality.
The steps outlined in this section are a logical progression of prac-
tice screening, evaluation, and selection for a particular watershed. Many
of these steps will require field work or contact with water quality mana-
gers who have access to estimates of loading concentrations and soil,topo-
graphic, and farm data. In addition, the analytical techniques, simulation
modeling, and linear programming described later in this manual may require
assistance from specialists. This forced interdependence between groups
working to improve water quality underscores the comprehensive and complex
nature of the task.
77
-------
SECTION 5
EXAMPLES OF THE AGRICULTURAL NPS SELECTION PROCESS
PURPOSE OF CASE STUDIES
This section demonstrates the methodology presented in Section 4. The
areas selected are from both non-irrigated and irrigated agricultural
regions. Entire watersheds, sub-watershed areas and farms were modelled.
In each case combinations of control practices were evaluated with respect
to total pollutant load reduction and associated control costs.
Different control strategies were developed for each case. Although
the techniques used to optimize levels of pollutant control and select
treatment areas differed somewhat for each case study, the objective of the
practice evaluations was to develop cost-effectiveness curves, as outlined
in Section 4.
CASE STUDY I: HONEYCREEK, OHIO
Description of Watershed (Step 1)
The Honeycreek Watershed is within the Sandusky River Basin, located in
north central Ohio (Figure 20). The watershed 'drains into Lake Erie.
Considerable interest has been expressed in the watershed because of the
deterioration of water quality in Lake Erie over the past few decades.
The Honeycreek watershed comprises 47,144 hectares of which over 80%
are cropped. Because of this intensive agricultural use and due to its
soils, Honeycreek can be considered representative of the glaciated, once
forested areas of the cornbelt. Honeycreek soils are generally of fine
texture with somewhat poor natural drainage on slopes of less than 3% (see
Table 7). The distinctive soils and topography of the watershed are a
result of glacial activity, and include areas of glacial drift, outwash,
ground and end moraines, in addition to lacustrine sediments, flood plain
alluvium and organic sediments. As shown in Figure 20, the watershed can be
subdivided into four sub-watershed areas (A, B, C, D) based primarily on
these physiographic characteristics. Subwatershed A is composed primarily
of glacial drift and dissected ground moraine while the soils in B are
derived from ground moraine (dissected and undissected) with glacial outwash
terraces. Much of the soils in subwatershed C are ground moraine
(undissected) and flood plain alluvium. Subwatershed D is primarily ground
moraine (dissected) and end moraine. The watershed has been subdivided into
18 distinctive sub-areas. Table 8 lists land use and physical
characteristics for each of the four main subwatersheds (A-D).
78
-------
Figure 20. Honey Creek watershed.
Problem Identification (Step 2)
Agricultural land uses In Honeycreek and other watersheds draining Into
Lake Erie contribute organic matter, nutrients and sediment. The water
quality problem most often cited in Lake Erie is the excessive growth of
algae which has been attributed primarily to phosphorus loading. The annual
production of algae in Lake Erie has increased twenty-fold in sixty years
(Honeycreek Report, 1979). Algae growth can interfere with recreational
uses of the water when it occurs near shore areas and can deplete dissolved
oxygen supplies. In addition to these nutrient enrichment problems, sedi-
mentation of navigable channels in the lake results in significant public
expenditures for dredging.
Although there has been considerable study of the water quality of Lake
Erie, the precise relationship between nutrient enrichment of the lake and
agricultural runoff and sedimentation has not been determined. Thus speci-
fic target load reductions from watersheds including Honeycreek have not
been established. For this case study analysis, a wide range of pollutant
load reductions are evaluated.
79
-------
TABLE 7. SOIL AND TILLAGE GROUP DESIGNATION OF HONEY CREEK SOILS
Soil
Group
1
2
3
4
5
6
7
Soils in
Group
Bono
Lorain
Luray
Chagrin
Papakating
Lenawee
Millsdale
Belmore
Haney
Digby
Belmore-
Morley
Marengo
Toledo
Wallkill
Shoals
Pewamo
Pewamo-Urban
Gallman
Hennepin-Ale .
Haskins
Mi 1 ton
Tillage
Group
4
5
4
1
1
2
1
Condit
9
10
Carding ton
Glynwood
Bennlngton
Blount
Mori ey
Ri tchey
Tiro
1
2
Source: Becker andForster (1978) .
* Tillage Group 1 -- naturally well-drained soils where yield is expected
to be equal to or greater than conventional moldboard plowing.
Tillage Group 2 -- response equal to or greater than conventional
when surface or subsurface drained.
Tillage Group 3 -- poor internal drainage is not improved by artificial
drainage measures so significant yield decreases are expected with reduced
tillage systems.
Tillage Group 4 -- these soils may yield less with no-till practices
even when surface or subsurface drained. Unlike group two these soils
have a high organic matter content which will decrease the yield varia-
tion between extremely wet f> dry years as compared to the expected
yield variation usiny a conventional plowing system.
Tillage Group 5 -- includes organic soils, recent alluvium strip mined
land and certain fine textured soils. Experience with changes in tillage
practice has not been sufficiently documented to determine the general
response by these soils to no-tillage.
80
-------
TABLE 8. SOIL AND LAND USE CHARACTERISTICS, HONEYCREEK WATERSHED
SUBWATERSHED
Land Use
Row Crops
Field Crops
Other Agr.
Forested
Other
Slope
Categories
0-2%
2.1-4%
4%+
Soil Drainage
Well & Mod.
Well
Somewhat
Poor
Poor & Very
Poor
Not Available
A
HA* (%)**
2964
2080
124
864
428
6460
4336
1104
1020
6460
1560
3968
732
200
6468
45.9
32.2
7.9
13.4
6.6
100.0
67.1
17.1
15.8
100.0
24.2
61.4
11.3
3.1
100.0
B
HA* (%)**
6044
3556
52
1372
676
11700
8512
2100
1088
11700
1820
7660
1924
296
11700
51.7
30.4
.5
11.6
5.8
100.0
72.7
18.0
9.3
100.0
15.6
65.5
16.4
2.5
100.0
C
HA* (%)**
9668
5420
124
1992
1352
18556
15360
2448
748
18556
848
12732
4632
344
18556
52.1
29.2
.7
10.7
7.3
100.0
82.8
13.2
4.0
100.0
4.6
68.5
25.0
1.9
100.0
D
HA* (%)**
4880
3516
72
852
1104
10424
8236
1556
632
10424
1348
4204
4648
224
10424
46.8
33.7
.7
8.2
10.6
100.0
79.1
15.0
5.9
100.0
12.9
40.3
44.6
2.2
100.0
*(HA) = Hectares
**(%) = Percent of Subwatershed
81
-------
Determining Applicable Control Measures (Step 3)
The management, vegetative, and structural control measures listed in
Section 3 were screened to identify measures which would be appropriate for
the designated problem, and compatible with the soil and crop/livestock
systems of the watershed.
Any measure that affects phosphorus movement is a possible candidate
measure since phosphorus has been identified as a limiting nutrient in algae
production in the lake. Both soluble and adsorbed forms of phosphorus are
known to contribute to the pollution problem. Because in-stream and lake
transformations of phosphorus are difficult to trace or predict, the rela-
tive contribution from soluble and adsorbed forms and any equilibrium rela-
tionship between the two forms have not been determined. Therefore, both
overland flow and soil erosion/sedimentation pathway controls were evalu-
ated. The effect of control practices on substances leached from the root
zone is also reported because the control of overland flow may contribute to
excessive leaching of non-adsorbed nitrate nitrogen.
Poor drainage of many Honeycreek soils limits the application of a
number of the candidate measures listed in Section 3. Some of the practices
which are of limited potential use in the watershed are listed below;
limitations are discussed.
Improved Timing of Field Tillage Operations (NIA-4)4--
Many of the heavier soils in Honeycreek are fall plowed because it
allows winter freeze-thaw cycles to help break up the soil and improve early
spring drainage. In fact this fall plowing may be accompanied by a network
of "dead furrows" which can increase surface drainage.
Using Mechanical Weed Control Methods (NIA-8) --
Most of the farming in the watershed is cash cropping with a distinct
trend to larger units (Venice Township, 1976). Since these types of opera-
tions have a high opportunity cost for labor, herbicide treatment of weeds
is used.
Reduced Tillage (NIA-9) and No Tillage Systems (NIA-10) --
Yield responses to reduced tillage are strongly influenced by soil
drainage, degree of tillage, and mulch cover. Poorly drained soils in
colder climates tend to decrease no tillage yields as compared with yields
obtained using coventional moldboard plowing. A number of soils in
Honeycreek would probably not respond well to no-tillage practices. Table 8
groups Honeycreek soils according to expected tillage responses (Triplett et
al. 1973).
Table 2 for description of practices.
82
-------
Contour Farming (NIA-11) and Contour Strip Cropping (NIA-15) --
Although contouring is possible on most soils in the watershed, it
tends to aggravate drainage problems where they exist. Also the complex
slopes in many areas of the watershed make contouring difficult and decrease
the efficiency of field operations.
Sod-Based Rotations (NIA-13), Permanent Vegetative Cover (NIA-16), Field
Borders (NIA-17) and Buffer Strips (NIA-18) --
Because cattle and dairy operations in the watershed are of minor
importance, the uses and markets for hay crops are limited. In addition,
the trend has been to increase corn relative to hay in most feed rations
which further decreases the need for hay.
Terraces (NIA-19) --
Both complex topography and poor soil drainage discourage the use of
terraces.
Choosing The Unit of Analysis (Step 4)
The Honeycreek watershed has been conveniently divided into four
physiographic regions. These areas are independent drainage units, thus
water quality management can focus on each sub-watershed. The following
study analyses effects of applying control practices to both the entire
watershed and to each sub-watershed area. The objective in choosing these
areas is to demonstrate the difference between uniform and non-uniform
levels of control within the watershed (see Appendix F).
The unit of analysis could be broken down even further to the farm or
field level. Case study II, the Yakima River Basin, will demonstrate this
approach.
Establishing the Base Condition (Step 5)
Establishing the base condition has both physical and economic dimen-
sions. Base physical conditions involve the variables described in Section
III:
a. Precipitation and Temperature
b. Topography
c. Irrigation Requirements
d. Soil Type
e. Cropping Practice
f. Stream/Lake Characteristics
Each of these parameters and its effect on base pollutant losses are
discussed below. Data sources and alternative ways of collecting the data
are also given.
83
-------
Precipitation and Temperature --
The Cornell Nutrient Simulation (CNS) model used in this case study to
estimate overland flow requires daily precipitation and temperature data.
Data used for Honeycreek were taken from a data file with information for 27
climatical areas in the eastern United States (refer to Appendix F). Figure
21 shows these regions. For each region the data included:
1. Average monthly precipitation and number of days with
precipitation .01" .
2. Average January and July air temperatures.
3. Average dates of planting and harvest of principle crops
in each region.
Figure 21. Climatological areas in Eastern United States with
similar precipitation and temperature patterns.
The data for Honeycreek, included in Region 17, is presented in Tables
9 and 10.
84
-------
TABLE 9. PRECIPITATION AND TEMPERATURE DATA, HONEYCREEK WATERSHED
(REGION 17)
Month
January
February
March
April
May
June
July
August
September
October
November
December
Total
TABLE 10.
Precipitation Number of Days Average Temperature
(cm) >.0254 cm °C
6.45
4.50
8.03
8.26
8.74
9.04
8.33
7.11
7.80
6.65
6.15
5.74
86.89
, CROP DATA FOR
14 -3.2
11
13
13
12
11
10 23.1
8
7
8
10
12
129
HONEYCREEK WATERSHED (REGION 17)
Crop
Barley
Corn
Hay
Oats
fsp.)
Rye
(fall we)
Soybeans
Winter
Wheat
Planting
Dates(P)
(Spring) No Appl
5/1-6/15
5/15-5/25
4/1-5/10
9/10-10/20
5/10-6/20
9/10-11/T5
Average Average
Harvesting Planting Harvesting
Dates (H) Dates (P) Dates (H)
i cation
10/10-11/30 5/25 11/5
9/5-10/5 5/20 9/20
7/10-8/5 4/20 7/25
6/20-7/15 10/1 7/1
9/30-10/30 6/1 10/15
6/30-7/25 10/5 7/15
85
-------
TABLE 11. PROPERTIES OF HONEYCREEK SOILS
Soil Texture Hydrologic
Group # Class Group
1
2
3
4
5
6
7
8
9
10
silty clay D
loam
silt loam C
silty clay C
loam
loam B
loam B
loam C
loam, C
silt loam
silt loam 0
silt loam C
silt loam C
Depth (ft) . ....
To Seasonal permeat>ili ty
High W.T.
0 - h .6-2
2-4 .6-2
floods
0 - h .6-2
> 6 2. - 6
1% - 3 .6-2.
3/4 - 2 .6 - 4.5
4 to > 6 2. - 6.
0 - h .6-2
lh - 3 .6-2.
f - 14 .6-2.
Available „ ,, _ , . ., .,
PH Water Bulk Total Available
Capacity Density P P
IB/ft 3 lb/ft3 ug/g ug/g
6.6
7.3
6.1
7.6
5.2
7.3
5.6
7.3
5.6
7.3
5.4
7.3
5.6
7.3
5.1
6.5
5.6
7.3
5.1
7.1
.17 - .22
.20 - .24 1.5
.17 - .22 1.5
.14 - .18 1.5
.14 - .18 1.7
.16 - .18 1.7
.14 - .18 17
.17 - .21 1.4
.17 - .21 1.6
.17 - .21 1.7
ug/g ug/g
685
848.08 23.90
455.38 15.87
Topography --
Estimates of slope and slope length for the Honeycreek watershed were
made using Conservation Needs Inventory (CNI) estimates (CNI, 1967). These
estimates for the Honeycreek watershed were improved by an expanded sample
and field verification during Lake Erie Wastewater Management Study (LEWMS)
activities (Stem, 1978).
Appendix F provides estimated slope and slope length for the Honeycreek
case studies and describes other methods of estimating slope and slope
length, data sources and examples of each approach.
Soil Type --
In the Honeycreek watershed many of the 44 soil types have similar
credibility, expected corn yield, and drainage characteristics. Table 11
lists the ten groups which were used and the principle soil series in each
group; properties of each of these ten soil groups are indicated. These
properties and values were used in simulating nutrient losses.
The principal criteria used in grouping soils were drainage character-
istics and response to no-till planting. Much of the information included
in Table 11 was extracted from county soil surveys. The number of soil
groups and criteria for grouping were influenced by the types and number of
practices evaluated.
86
-------
Once the soils were grouped it was necessary to determine the amount of
land in each soil group for each watershed. Since county soil surveys do
not compile data by drainage areas, this information is generally not avail-
able. The least time-consuming method of making this estimate is by grid
sampling from soil maps. In the Honeycreek watershed 500 single hectare
cells were sampled to determine the incidence of each soil type (Table 12).
The number of grid cells sampled and their size can vary. The criteria used
in determining the number of cells to be sampled and some examples of data
derived for Honeycreek are included in the Appendix F.
TABLE 12. HECTARES OF CROPLAND IN EACH SOIL/SLOPE GROUP FOR EACH
SUBWATERSHED
Soil
Group
1
2
3
4
5
6
7
8
9
10
Other
Slope*
Category
1
1
1
1
2
1
2
1
2
1
2
3
1
1
2
3
1
2
3
Subwatershed
A
63
583
157
94
274
58
72
422
63
-
5
143
368
58
305
313
1144
2127
_
211
B
198
383
60
5
8
5
_
13
18
77
288
-
1527
_
259
310
4031
3733
190
597
C
737
835
158
24
71
-
-
228
187
-
_
-
2913
_
512
6
8224
4001
_
662
D
1451
33
868
-
_
_
-
_
_
-
_
-
830
_
1059
185
2040
1351
_
2606
Total
2249
1834
1243
123
353
63
72
663
268
77
293
143
5638
58
2135
814
15439
11212
190
4076
47T43
*Slope Categories: 1 = 0-2%; 2 = 2-4%; 3 = 4+%
Cropping and Livestock Systems --
Knowledge of crops grown, common rotations and livestock systems are
critical to understanding both the economic structure of farms and soil/
water movement on cropland. Vegetative cover and crop residues strongly
influence erosion and runoff potential.
87
-------
No published data exist from which crops grown on specific soil manage-
ment groups can be determined. For any given year, field sampling could be
used. (This approach would be particularly attractive if slope length were
to be determined in the field also.) The crop grown on a particular field
or the rotation planned for a field, however, is subject to change. In
livestock areas where feed rations are not changed greatly from one year to
the next, crop acreages will not vary much. In cash crop areas similar to
those near Honeycreek, annual decisions made on crop acreage are based on
previous year's prices and expected changes in market conditions during the
next year. These price variables make crop acreages in cash crop areas
difficult to predict, so only rough approximations of expected crop acreages
can be made.
A realistic portrayal of crop acreages in the watershed can be based on
county statistics showing the ratio of crop acreage to total production in
the county. Table 13 shows the livestock estimates derived from published
agricultural statistics.
TABLE 13. NUMBER OF LIVESTOCK IN HONEYCREEK WATERSHED
County Beef Cattle and Calves Milk Cows and Heifers Hogs
Crawford
Huron
Seneca
2778
441
3883
7102
410
132
850
1392
4142
760
5965
10867
In the Honeycreek area a survey was recently conducted in a represen-
tative area of the watershed (Venice, 1976). Data from this survey were
used to determine approximate acres of each major crop, and also livestock
numbers (see Appendix F). This type of information is not available in most
watersheds but a survey to collect these data can be performed by project
personnel. Assistance can also be gained by contacting coopera- tive and
county extension, ASCS and SCS offices.
Stream/Lake Characteristics --
The relationship between agricultural nonpoint source loads and in-
stream or lake pollutant concentrations is often unclear. As discussed
in Section 2, this missing link in water quality management generally does
not allow precise estimates of the water quality benefits which will result
from nonpoint source controls.
Analysis of alternative practices in the Honeycreek watershed focuses
on the decrease in edge-of-field phosphorus, nitrogen and sediment loads.
Because the watershed is intensively farmed it is assumed that control of
these pollutants will decrease their concentrations in the stream. However,
the precise nature and extent of the change is not now known.
88
-------
Previous LEWMS research efforts in the Honeycreek watershed have
resulted in the collection of a significant record of water quality data
(Water Quality Data, 1978). Although these data do not solve the problem of
relating agricultural nonpoint source controls to lake or stream quality
improvements, they are useful for problem identification and hydrologic
model calibration and validation.
Farm Economic Structure --
For the Honeycreek region, the costs and returns associated with live-
stock and crop production were estimated (Tables 14 and 15). Production
costs included both fixed and variable costs. Capital investments, such as
machinery purchases, are examples of fixed costs; labor, feed and fuel are
variable costs. A detailed discussion of all cost estimates is found in
Appendix F.
TABLE 14. DAIRY BUDGETS FOR THE HONEYCREEK WATERSHED
Variable Costs
Cone. Protein (SB)
Minerals
Salt and Dicol
Milk and Starter
Sub-Total Feed Costt
Vet. and Med.
Breeding (DHI)
Utilities
Bedding
Mi sc. and Suppl ies
Market Costs
Interest on Operating Capital
Total Variable Costs
Fixed Cost
Labor
Interest
Cow/Calf Replacement§
Total Costs
Da i ry*
Cows
($)
41.00
14.00
4.00
59.00
24.00
27.00
26.00)
20.00)
20.00)
72.00)
22.00
270.00
270.00
69.00
215.00
554.00
824.00
Da i ry t
Replacement
($)
138.00
6.00
2.00
36.00
182.00
15.00
25.00
15.00
47.00
284.00
135.00
95.00
100.00
330.00
614.00
*Production 13,000 #/cow, receipts $1683/yr
tReplacement period birth to freshening, 36 mo.
tCorn/hay feed requirements accounted for separately as part of farm group
production
§Cow replacement = 0.35 x $614
89
-------
TABLE 15. SOYBEAN BUDGETS FOR THE HONEYCREEK WATERSHED
Soybean
Conventional
($)
Soybean
Minimum
($)
Soybean
No-Till
($)
Variable Cost
Seed
Chemicals
1
Fuel, Oil, Greasy
Repair and Misc.'-
Labor2
Sub-Total Operating Cap.
Interest on Operating Cap.
Trucking^-
Fertilizer5
Fixed Cost
Management Costs
13.00
8.00
21.00
15.00
57.00
2.71
13.00
7.36
19.32
13.80
53.48
2.54
13.00
7.20
18.90
13.50
52.60
2.50
Machinery Depreciation
TOTAL
Chemicals
Lorox
Lasso
30.00
89.71
7.50
7.50
27.60
83.08
7.50
7.50
27.00
82.10
7.50
7.50
Chemical - cost added in LP and interest
p
Source: N. Rask and D.L. Forester, "Corn Tillage Systems - Will Energy Cost
Determine the Choice", Agriculture and Energy. Academic Press, Inc.
1977. (Base price = Conventional till, Base price x .92 minimum
till, Base price x .90 = no-till).
Interest on Operating Capital - 9.5% for 6 mo. = .0475
Trucking cost = $.01 per bu. Cost added in LP
Fertilizer function of yield. Cost added in LP
P905 (Ibs) = 26.06 + .555 (yield) - .355 (P Test)
KgO (Ibs) = 80.556 + 1.333 (yield) + .75 (CEC) - .33 (K Test)
Management - 29.5<£ per bu. Cost taken out in LP
90
-------
Evaluating Control Measures (Step 6)
Cost and effectiveness of control measures were the major basis of
evaluation. Initially, practices were screened for suitability in terms of
physical and economic limitations and in terms of compatability with the
present farm systems of the Honeycreek watershed.
Effectiveness --
Effectiveness was determined by comparing edge-of-field gross pollutant
loading after practice implementation to that loading occurring under the
base condition. No effort was made to estimate pollutant loadings entering
the water body itself. This would require assessing the transformations of
pollutants as they move from the field to the stream which would be
extremely difficult. Once pollutants moved from the field they were assumed
to reach the receiving water.
Field losses were estimated using the Universal Soil Loss Equation
(USLE) for gross soil erosion, and the CNS model for average annual overland
flow and nutrient losses. The commonly used USLE (Wischmeier and Smith,
1978) estimates average annual soil erosion based on a regional rainfall
factor, the intrinsic erodibility of the soil, slope and slope length of a
given field, the crop management system, and soil conservation support prac-
tices.
The Cornell Nutrient Simulation (CNS) Model developed by Haith and
Tubbs (1978) consists of a hydrologic subroutine to calculate runoff and
leaching losses and a loading subroutine to estimate losses of adsorbed
phosphorus (P) and nitrogen (N), as well as losses of dissolved P and N to
surface and subsurface flows. The CNS model computes daily soil moisture
concentrations and monthly soil nutrient budgets by quantifying all signi-
ficant measurable inputs and outputs of water, nitrogen and phosphorus.
Daily precipitation and temperature data are replaced by probabilistic
inputs corresponding to the appropriate geographic area. Runoff is cal-
culated from these temperature and precipitation inputs by use of a modified
form of the SCS Curve Number Equation. It is assumed that soil drainage is
not limited by either a high water table or impermeable layers.
In the CNS model, two soil layers are modelled separately: a top layer
of soil 10 cm thick and the 20 cm soil layer directly below this top layer.
Organic nitrogen and total P are modelled for the top layer; inorganic
nitrogen is modelled for both layers. Runoff losses, which include surface
and subsurface losses, are assumed to occur from only the top layer of soil
whereas percolation losses take place within both layers.
Nitrogen losses in runoff and percolation are calculated as the product
of concentration and water volume, where concentration is determined from
the average nitrogen level in the soil during the month. Because this model
assumes that no losses of N via ammonification or denitrification occur
after application, it will over-predict N losses in runoff or percolation
over the long term. However, adjusting the initial fertilizer input to
compensate for ammonification and denitrification losses may yield more
accurate results.
91
-------
Phosphorus losses in runoff are modelled in much the same way as N
losses except that the adsorption coefficient (Ka) of the soil must also be
considered. The coefficient affects soluble phosphorus availability. Phos-
phorus losses are typically underpredicted (particularly for intense storms)
since the model does not account for the increase in quantity of soluble P
associated with an increase in water volume.
The CNS model is applicable to any land area. It does not develop pol-
lutant delivery ratios nor does it estimate in-stream quality parameters.
However, it is a useful tool in comparing pollutant pathway control as a
result of the implementation of nonpoint source control practices.
Cost --
In addition to the cost to install and maintain control measures, costs
should include reductions in farm income resulting from investment in water
pollution controls. While other industries are able, at least partly, to
pass on cost increases to consumers, individual farm operators cannot (Nicol
et al., 1974). For this reason it was important to look closely at how pol-
lution control influences a farm's cost structure. These changes in cost
structure may mean appreciable decreases in income for certain farming
regions or even place farm producers at a serious competitive disadvantage.
The method used for estimating the income effects of control practices
was linear programming (LP). Linear programming methods are particularly
helpful in the evaluation of nonpoint source controls since both single
practices and sets of practices can be evaluated. In addition, constraints
can be established to limit runoff or erosion in certain subwatersheds or
soil groups as well as the watershed as a whole.
A base LP solution maximizes or minimizes an objective function under
certain restrictions (constraints) on the amount of land, labor and capital
available to the farmer. For this study, the objective function was to
maximize net farm income. The options for use and management of land are
referred to in this analysis as 'activities'. The activities chosen for the
LP model of the Honeycreek region reflected realistic alternatives available
to the farms in the area. The number of choices was minimized since
consideration of too many options creates an unwieldy array of alterna-
tives. Net farm income was maximized by varying three sets of activities
(management, vegetation, and structural controls) according to the following
equation:
Y = k 1 m n (Aklmn - Nklmn)
Where: Y = net farm income
k = combination of soil group and slope. Ten distinct
soil groups were considered and three slope
categories (0-2%, 2-4%, 4%+).
1 = tillage practices: conventional moldboard, chisel
plowing and no-till planting.
92
-------
m = crops: corn grain, soybean, wheat and meadow.
n = conservation practices: contour farming, crop
rotations and diversion waterways.
Aklmn = hectares of land with conservation practice (n)
applied to crop (m) and field type (k) using
tillage system (1).
Nklmn = annual net revenue after producing one hectare of
crop (m) for a particular combination of field type
(k), tillage method (1) and conservation practice (n).
Constraints on sediment and nutrient losses can be expressed as:
(Aklmn - Pklmn) <^ loadklmn
Where:
Pklmn = pollutant loss (kg/ha) for a specific combination
of field type, crop, tillage and practice applica-
tion.
loadklmn = residual losses of sediment or nutrients for a
particular combination of k, 1, m, n.
Linear programming was used to compare the relative economic efficiency
of alternative nonpoint source controls. These calculations were performed
without consideration of cost-sharing. No changes in the price received or
paid by farmers were assumed with the implementation of nonpoint source
controls. In each case partial budgeting methods were used to determine net
return (see Appendix F).
Developing an Optimal Control Strategy (Step 7)
There are a wide variety of control options ranging from traditional
soil and water conservation structures, such as terraces and diversions, to
changes in tillage systems, fertilizer application rates or crop rotations.
In order to Insure that a wide range of practices was considered, while at
the same time minimizing the nunber of alternatives, practices were selected
from each of the three categories mentioned earlier—management, vegetative
and structural. Management options included changes in tillage practice and
contour farming. Vegetative controls allowed one of four crops rotations to
meet soil loss restrictions. Structural alternatives were reflected by the
use of diversions to reduce slope length and thus, meet erosion restric-
tions.
To establish the relative improvement by a control practice set, it was
compared to the base condition. Both loading estimates and income levels
were calculated for the base conditions. It was assumed that for the base
condition, those cropping systems which maximized profits would be
selected. As pollutant loss constraints were placed on the watershed, the
objective function continued to maximize profits using nonpoint source con-
93
-------
trols to reduce pollutant losses. These new income levels were compared to
the base solution to determine pollutant control costs.
The linear programming model and the estimates for overland flow and
leaching were linked in the following manner. The quantity of land in each
soil management group having a certain tillage system, crop and conservation
practice was determined for the base condition and for increasingly strin-
gent levels of soil erosion control using linear programming methods. Total
runoff, nitrogen and phosphorus losses were then estimated using the simula-
tion model for each combination of activities at all levels of erosion con-
trol. The results indicated the change in farm income, associated reduction
in gross soil erosion, reduction in runoff losses, and the associated losses
of fixed and soluble nutrients.
Results --
Figure 22 illustrates the change in income with application of the
three different control sets to meet soil loss restrictions in subwatershed
A. A base solution for each linear programming (LP) run establishes the
profit maximizing case with no soil loss restrictions. Base solutions dif-
fer depending on the type of tillage system assumed. Since reduced tillage
systems greatly reduce erosion and have a relatively low cost they become
part of the profit maximizing base solution for managerial controls.
From Figure 22, it is apparent that management controls, particularly
changes in tillage practices, are a very efficient control strategy. At
increasing levels of erosion control, vegetative and structural controls
reduce income appreciably.
Figure 23 indicates how different management practices entered LP
solutions. As soil erosion is limited, the area in subwatershed D which is
treated with various tillage options changes. Both spring plowing and con-
touring enter the LP solutions at relatively high erosion limits while
no-tillage enters at relatively low limits. Thus spring plowing and
contouring are relatively more cost-effective than no-tillage in this case.
However, in other watersheds with well drained soils one would expect the
opposite.
Vegetative controls appear the most costly alternative since increased
areas of wheat and meadow were needed to meet soil loss constraints (Figure
24). These lower value crops, although effective erosion control measures,
result in appreciable decreases in return as shown in Figure 22. Struc-
tural controls5 have a relatively high capital cost and result in some
cropland taken out of production. Thus, although high value crops (corn and
soybeans) can continue to be grown, this advantage is offset by capital
charges and decreases in total production. Figure 25 shows the area of
cropland with diversions as erosion constraints become more stringent.
5A1though only evenly spaced diversions (or "cross-slope drains") were
considered, these controls act like graded terraces and are representative
of practices which reduce slope length.
94
-------
Figure 22. Relationship between gross farm income and soil loss
constraint levels for three erosion control strategies,
320
260
240
Z
or
(T 200
160 -
120
EROSION CONTROL TECHNIQUE
a Vegetative
• Structural
o Managerial
I
12 10 8
SOIL LOSS, MT/ha
Figure 23. Management practices entering linear programming
solutions - Subwatershed D.
100
80
60
Q.
O
£ 40
u.
o
zo
~~~~ Spring plow
Foil plow
— — Minimum tillage
No tillage
_ o — Contour
SOIL LOSS, MT/ho
95
-------
Figure 24. Linear programming solutions giving cropping areas
for different soil loss constraint levels -
Subwatershed A.
so r
CORN
14 10
SOIL LOSS, MT/ho
Figure 25. Linear programming solutions giving cropping areas
for different soil loss constraint levels -
Subwatershed A.
100 r
\6 16
14 12 10 8
SOIL LOSS, MT/ho
642
96
-------
Runoff reductions are shown in Figure 26. They follow a trend similar
to that of soil loss reductions: as control increases (i.e., return
decreases) runoff decreases, with vegetative practices being less effective
than management controls. Nutrients transported either with sediment or in
surface runoff are effectively treated by both management and vegetative
practices although management options are the least costly (Figures 27 and
28).
Figure 26. Changes in income for different levels of runoff controls.
336
272 -
192 -
112
Management
Vegetative
j I
I I
20
18
12
10
RUNOFF, I08m5/yr
The simulation model used to estimate runoff was not sensitive to the
implementation of diversions so they were not evaluated as a management
option.
Soil loss limits can be placed on the entire watershed as well as
individual watersheds to allow for greater planning flexibility. In the
latter case, for the same average soil loss, returns for a specified sub-
watershed can be greater or less than those returns received when it is
individually constrained. Figure 29 shows the variation in return for the
four sub-watersheds at different soil loss constraint levels. One would
expect this variation to increase appreciably for smaller sub-watershed
units or where crop and livestock activities differ more.
97
-------
Figure 27. Changes in income for different levels of dissolved
phosphorus control.
336 r
272
z
o:
3
Manogamtnt
Vegetative
1100
1000 900 800
DISSOLVED P IN RUNOFF, kg/yr
700
Figure 28. Changes in income for different levels of solid phase
nitrogen control.
336 r
272
u
K
192
112
• MMogommt
'Vogttatlvi
I I I I I I I I I I I I I I I
ISO
100 50
SOLID PHASE N, MT/yr
98
-------
Figure 29. Control cost variations among subwatersheds.
UJ
-------
Figure 30. Changes in income for different levels of dissolved
nitrogen control.
336 r-
272
LJ
at
192
112
Monogtrlol -
/ V«g*!otiv«
I
_L
I
I
_L
I
750 700 660 600 550
DISSOLVED N IN PERCOLATION, MT/yr
500
As shown in Figure 30, those nutrients which percolate down through the
soil profile, specifically nitrate nitrogen, were increased when practices
designed to reduce surface losses were applied. With increasing soil loss
constraints, dissolved N losses increased especially for vegetative
controls. The implementation of the management options was less dramatic
since crop residues do not increase infiltration as much as a sod cover.
CASE STUDY II: YAKIMA RIVER BASIN, WASHINGTON
Two models were developed to evaluate control alternatives for the
Yakima River Basin - a farm model and a watershed model. The watershed
model characterizes the Yakima Basin in terms of seven subwatersheds. Agri-
cultural, hydrological, and water quality simulation submodels are used to
analyze the cost-effectiveness of control alternatives for the basin as a
whole. The farm model characterizes soils and cropping practices on a farm
scale and is used to evaluate the cost and effectiveness of different irri-
gation systems. Steps 1 and 2 below, Description of the Watershed and
Problem Identification, include information which is relevant to both the
watershed and the farm model. The rest of the evaluation process, Steps 3
to 7, is carried out for each model separately.
100
-------
Description of Watershed (Step 1)
The Yakima River Basin lies in south central Washington and encompasses
about 6,000 square miles (Figure 31). The basin contains approximately
450,000 acres of irrigated land, most of which was originally developed
through a series of Bureau of Reclamation irrigation projects. There are
many fruit orchards of less than 25 acres. The average farm size is less
than 200 acres.
Figure 31. The Yakima River Basin, Washington.
Kachess
Keechelus * f'Cle El urn
Lake
IRRIGATION AREAS
REACH TERMINUS
COLUMBIA
RIVER
Problem Identification (Step 2)
Water quality in the Yakima River generally has been excellent above
the city of Yakima. River nitrogen concentrations have remained below the
algae bloom level of .30 mg/liter during most of the year. Temperature has
usually remained below the Washington Department of Ecology Class A standard
101
-------
of 65°F. Sediment loads have generally been small except during periods of
heavy snow melt or precipitation. The Naches River has also generally met
water quality standards.
Approximately one-half of the 288,000 ha-meters (2.4 million acre feet)
of water diverted annually from the river is used above Yakima and one-half
below Yakima. There is a considerably higher per hectare use of water along
the upper reaches of the river. Below Yakima, water quality grows progres-
sively worse as a result of major irrigation diversions and large return
flow volumes. It is estimated that 80 to 90 percent of the water in the
lower reaches (6 and 7 of Figure 31) during the late summer months is return
flow from the upper reaches. These return flows carry a higher con-
centration of nitrogen and suspended solids than are contained in virgin
river water. Consequently, nitrogen concentrations exceeding 1.0 mg/liter
and water temperatures exceeding 80° are frequent during the summer. Tur-
bidity caused by suspended sediments and algal blooms afflict the lower
reaches during most of the irrigation season, with effects being most pro-
nounced between mid-July and mid-September when water flow is lowest and
irrigation use is highest. During this period, low water quality inhibits
fish passage and recreational activities, and turbidity causes problems with
downstream sprinkler systems.
FARM MODEL
Determining Applicable Control Measures (Step 3)
Existing water quality problems in the Yakima are associated with
nitrogen and sediment delivery to the Matches and lower Yakima basins. Thus
control of erosion and leaching pathways are of primary concern.
Because of the low irrigation efficiencies in the watershed, a number
of practices appear to be realistic control options in the Yakima.
— Cutback Irrigation System (IA-25) and Improved Water Management (IA-12)
This system employs additional labor to reduce stream size once the
stream has reached the field bottom, plus an improved level of management.
The improved management includes optimal timing of irrigation, and length of
irrigation set time. It is estimated that total water loss from this system
will be 50 percent of the loss from current irrigation practices.
— Tailwater Reuse System (IA-26) and Improved Water Management (IA-12)
This set of practices modifies the current system by reusing the tail-
water and employing an improved level of management as described for the
above irrigation system. It is estimated that total base water losses can
be reduced by 80 percent and that runoff losses will be negligible.
102
-------
-- Automated Tailwater Reuse System (IA-26) with Gated Pipe (IA-25) and
Remote Water Control (IA-12)
The practices included in this system are automatic reuse systems using
gated pipe and remote sensing devices to control stream flow and set time.
With these practices total water losses can be reduced by 80 percent from
the base situation, and runoff losses will be negligible.
Side-Roll Sprinkler System (IA-29) With Improved Water Management (IA-12) —
See discussions in Section 3.
Solid Set Sprinkler System (IA-29) Operated with Improved Water Management
(IA-12) -
See discussions in Section 3.
The farm model developed for the Yakima was also used to evaluate less
traditional policy options. Although some of the options considered are
contrary to existing legal and institutional practices in the watershed,
they do provide a means of observing optimum adjustments to meet water
quality standards (see Section 3, and Appendix B). Although some of these
options are not practical at this time,they should be considered as possible
future measures which could evolve over a long-term to meet water quality
standards in a cost effective manner.
The policy options evaluated by the farm model include:
1. Constraining total nitrogen outflow from the farm.
2. Placing a per unit tax on nitrogen losses leaving the farm.
3. Making payments to the farmer for nitrogen and sediment
pollution abatement.
4. Constraining total sediment outflow from the farm.
5. Placing a per unit tax on sediment outflow from the farm.
6. Constraining the total use of N fertilizer on the farm.
7. Constraining the total use of irrigation water on the farm.
8. Placing a per unit tax on N fertilizer applied.
9. Placing a per unit tax on water applied.
10. Paying the farmer for every hectare of close-growing crops
produced.
The first five policies are administratively or legally infeasible at
present, unless all return flow waters can be collected by a common drainage
system and monitored. The policies can still be used, however, to determine
a norm to which other policies can be compared. Taxing or constraining
sediment loss remains potentially feasible, but difficult to implement for
individual farms.
Choosing the Unit of Analysis (Step 4)
The farm model is convenient for demonstrating the effectiveness and
cost of a number of different irrigation systems. The differences in soils
and cropping practices in the Yakima are evaluated by sensitivity analyses.
103
-------
Establishing the Base Condition6 (Step 5)
The more than 2,000 farms in the basin are organized into several
independent irrigation districts. Existing water rights in the basin have
normally been sufficient for crop production. Thus, few incentives have
been provided by water costs or supplies to improve irrigation efficiency
through better management and/or capital expenditures (a notable exception
being the drought period of 1976-77, when several irrigation districts were
programmed to receive only enough water to keep perennial crops and a few
annual crops alive). Consequently, surface or furrow irrigation is used on
60 to 70 percent of the irrigated land. The remaining acreage is irrigated
by sprinkler irrigation methods adopted primarily in response to growing
labor shortages and higher wages. A large proportion of the tree fruit
acreage is irrigated with solid set sprinkler systems.
While some 80 different crops are commercially grown in the Yakima
Basin, the economic returns and hydrologic characteristics of the basin can
be represented adequately by 6 to 10 crops. A typical farm would grow no
more than 3 to 5 crops.
Regions 1 and 2 (Figure 31) are dominated by forage crop production
which supports an extensive livestock industry. In Region 2, small grains,
sugarbeets, and vegetables are also grown. Region 3 is a relatively small
area above the city of Yakima in which diversified crop production pre-
vails. Region 4, lying along the Naches River, is dominated by sprinkler
irrigated tree fruits. Regions 5, 6 and 7 are the largest and most inten-
sively cultivated areas in the basin. Level land, rich deep soil, and
favorable weather conditions make these regions suitable for almost all
irrigated crops. Relatively large tracts of less productive land are never-
the less used for forage crops and pasture.
Gossett (1975) developed a linear programming model (see Appendix B) to
represent an individual model farm in the Yakima River Basin. The model
predicted the seasonal quantity of nitrogen leaving the farm in subsurface
drainage and the amount of sediment carried by surface runoff. These losses
were predicted for each of 5 crops, 4 fertilization rates, and 6 irrigation
systems. Each crop therefore could be produced 24 possible ways.
A total of 560 acres of gently sloping land and a deep silt loam soil
was assumed for the model farm. The crop mix included potatoes, sugarbeets,
wheat, spearmint, and alfalfa. The farm model established a base irrigation
system consisting of conventional surface irrigation methods using open
ditches, siphon tubes and furrows. The model was then used to evaluate
nitrogen and sediment losses for each of the control and policy options.
Control alternatives were evaluated with respect to the effectiveness of the
measures in controlling pollutant pathways and the farm costs associated
with these controls. Costs included expenses for land and operator labor.
Constraints were imposed to control crop rotations while maximizing net farm
revenue for a single production year.
6Physical and economic details of the watershed can be found in Gossett
(1975).
104
-------
Evaluating Control Measures (Step 6)
The capital and annual operating cost of the base irrigation system and
the five alternative control measures are summarized in Table 16.
Deep percolation loss was assumed to be a constant percentage of the
amount of water applied for the base system and option a. It was assumed to
be the total loss for systems b and c, which used a form of pump-back irri-
gation. Deep percolation under sprinkler irrigation was assumed to equal
the total loss minus 10 percent of water applied, which represents the loss
to wind and evaporation. Table 17 shows the estimated irrigation effi-
ciencies and deep percolation losses which were used in the farm model.
A base solution was obtained (see Table 18) when the farm was irri-
gated by the existing system, free of contraints. The only internal con-
straints were restricted crop acreages in order to represent realistic crop
rotations. These constraints were: potato acreage 32 ha (80ac.); sugarbeet
acreage 57 ha (140 ac.); alfalfa acreage 40 ha (100 ac); and spearmint acre-
age 57 ha (140 ac.).
When sediment and nitrogen outflows were unconstrained, the 560 acre
farm produced a net revenue of $87,095 or $155.53 per hectare (Table 18).
Total nitrogen leached through the root zone was calculated at 58.9 Kg/ha
(52.5 Ibs. per acre), producing a nitrogen concentration in the subsurface
drainage water of 21.3 ppm. The base solution also had 1744 tonnes of sedi-
ment leaving the farm, or 7.69 mt/ha. A total of 160 cm of water was
applied per ha. of farm land, with resultant deep percolation and runoff
losses of 28.7 and 55.4 ha-cm., respectively.
The effects of alternate irrigation systems are compared with the base
solution in Table 19. These solutions imposed no constraints or economic
incentives to influence pollution abatement beyond those provided by the
irrigation system. The solutions thus emphasized the marginal value of
management, labor, and capital in reducing nitrogen and sediment waste
loads. Comparing the existing system with system a, one can see that
improved management would reduce nitrogen leaching by 25 percent and sedi-
ment yield by 62 percent at a total farm cost of $3,477, or $15.31 per ha.
This cost arises because the additional labor costs more than the savings in
fertilizer use. The 3,900 hours required to irrigate the 227 ha (560 acres)
with system a would require 3 to 4 full time irrigators. Operators would
generally adopt more capital intensive irrigation systems in order to reduce
required labor.
Surface tailwater reuse, system b, was able to further reduce nitrogen
leaching losses below those of system a, and to completely eliminate sedi-
ment discharge. The reduced nitrogen losses may have been an artifact
because deep percolation was assumed to be a constant proportion of the
amount of water applied. With this particular system, water remains in the
irrigation furrows for the same length of time as for the base siutuation,
even though total water use is less because of the "reuse" of normal runoff
losses. Total net revenue was increased slightly for this set of assump-
tions, because nitrogen fertilizer savings were greater than the increased
costs of the tailwater reuse system.
105
-------
LU
>
1—!
U-
LU
n:
Q
<
5:
1 1 1
f=
GO
•w
^~
GO
LU
GO
-, <
{-CQ
'2s
< -
Q
•Z. "
cC GO
LU
1— >
z: i— i
LU I—
2T<
GO ct:
LU LU
E> 1—
E: —i
-HCC
CAPITAL
CONTROL
•
•J3
LU
I
— J
CQ
<:
CD
CJD
•i— f
r- 4-> to
to to
3 S- 4->
C d) CO
E Q. O
C
r— £
4-> 00
••- O)
Q. >
to c
O 1— I
I?
CU
4->
CO
£>
c
O
4J
to
CD
•r™
s_
1 — 1
<& 10 CvJ
C3 O CO
LO i r*i /TV
l-I J Q^
1 — 1
n3
,c
^^
v>
« K S
00 CO ^
«— ' I—I ^j-
1
o a> i — ro
•I— > fQ
•*-* o c: -a
fTj C O ^-
a> S..2 o
•r— f~* i ^ ^T
, fc 4-1 Q_
« S- •!-•!-
CO S- "O • E
CU -r- S- T3 CU O
-C O O 3 CU O
CU CU X r- i- U
O-i — CU O 4-*
O C t/l /r<
CU E -r=
-E CO CU 2 jr S-
-»-> to 4-> « s £ 5;
•"— -O CO 4-> i- T- 4->
5 >, E S- s re
*O (/) rtj ^ •>
C E 4- C r^
O "O CJ CU O T-
•J- E -i- CDjii .,- ra
T~"^ *o (/) (O t-J -4~J *t~^
03 IO E IO IO
CD •> J3 to JO CD CU
T °° E 4J -^ co
J- CU CU =3 S- !3
i •§ E 5 § u i £
+J C^ -r- 0
CU E CO 4-> 4-> CU O
o E a> to to o +->
(O O CD CD S- to
M-^IfO OJ-r-O M-Q.
^-Q-C Ei-^ ^H"
~^ — *••- k— ~ I"- .ij >~ ^
3 -i-nj tos-to 33
GOtOE GO-F-i— GOO.
LU
GO
2 * -°
co r-» co
CTl CTl IV.
• • •
IO IV, r-l
LO •* CO
i 1
LO O
i— 1 CM
• •
CO LO CO
LO cn CTI
ro oo LO
i — i
E
OJ
4-5
CO
>, CO E E
co O> O o
0 .,- .,-
CL) •!- 4-> 4->
CO > fO tO
3 oj cn cn
CD "O -i— -r—
s- s- s-
CD S- S-
0 E .r- .,_
4-> CO 4-> 4->
tO C E c"
E CD CU CD
O CO i — , —
I \ ^ ^_
^ — .C i—
5 i if if
5T3 S S
c i- S-
c: to CL> cu
O i — i —
•r. X •* -^
4J Q. c: c
fO *r~ «i— «|—
cn a. s- j_
••-0.0.
J- T3 on oo
S- O) +J 40
•r- 4-> i— c 4-> C
tO i — CL) CD CU
O5 O E co E
Uc fl i tr:
*- CU 1 CD
to CD i Q> T3 CD
II C~ fit _w -
^ t- CD to •!— to
*- '•- "O C r— E
3 CO T- to O to
GO 3 GO E GO E
0 -0 OJ
to
4->
•r—
Q.
IO
O
CO
O)
"O
13
o
c
•r~
4->
•^
JO
CO
4->
CO
O
O
S-
CL)
0
Q.
CJ
•r^
S-
j i
T— '
O
O)
C
to a
2 4-
£Z
•\ >|_
e iri
^— '<_
0 E
J2
to ~a
i — C
to
c
O CO
•r- 4->
4J CO
IO O
CD O
£ C
S- O
•r— «r—
4J
CO to
CU N
T3 -r-
3 4->
< C
0 C
X E
LU to
to
)
>
)
,
CO
cu
>
cu
O)
0
•i—
s_
Q.
•vl-
r^
CTl
r— 1
4->
O
CU
i —
M-
2!
CO
i ^
•f~*
CO
o
0
_a
106
-------
00
OO
00
UJ
00
00
o
8
o:
LU
UJ
O
O
UJ
O Z
»—( I—I
H oo
O CO
a ce
LU
§
r— O 00 «t LO
t^ co i-- r— co
^J- lO Lf) CVJ I—
01 co o «t n
m i^ i^ ^o oo
o 10 c\j o cj
i— «a- ro ro vo
Ol r— •* CO LO
i— 00 VO CO f
oo o i^ «fr cn
OO C\J CO
CO CVJ 00 r— CVJ
CM UD LO CM O E
• CO GO CO CO Cft O i— r—
O
U
01 o> c\j n ro
l~* CM r-
rv. oo co
CD i— 00 113 LO IO
CD CM i— CM i—
O
r^ co r^ co
LO \£> vo LO r^
in CM oo LO m
CSJ CO O CO ^~
oo fo i-~ co ro
co
•M J->
« 01 C
CD 01 >f- IQ
O J3 E M-
+J 1- <-> C r-
+-> Dl Q) 0) **-
O 3 J= Q-r—
D- 00 3 00 «I
5
cn
3 J
00
1 01 <4-
0.1—
00 «C
107
-------
TABLE 18. BASE SOLUTION FOR THE 227 HA. MODEL FARM, YAKIMA RIVER BASIN
UNITS TOTAL PER HA
Net revenue
Nitrogen losses
Sediment yield
Nitrogen applied
Water applied
Irrigation labor
Deep percolation
Runoff
$
Kg
tonnes
Kg
ha-cm
hrs
ha-cm
ha-cm
87,095
13,337
1,744
63,265
36,297
1,990
6,533
12,580
155.53
58.82
7.69
279.04
160.12
3.55
28.72
55.41
A surface automatic reuse method, system c, would allow significant
reductions in nitrogen fertilizer, water, and labor use, while maintaining
constant production. More efficient use of inputs would allow system c to
be adopted at the relatively small cost of $3*026, or $13.34 per ha. This
is despite a required capital investment of at least $371 per ha., because
nitrogen leaching was assumed reduced to 48 percent of the base level with
this system. The net income position would be slightly less than the base
condition.
The side-roll sprinkler, system d, proved to be uneconomic. Similar
abatement reductions could be obtained at smaller private costs with other
irrigation systems. Adoption of system d would cost over $9,000, while
still producing 11,029 kg (24,319 Ibs) of leached nitrogen. The solid set
sprinkler, system e, could reduce leached nitrogen by 50 percent, but only
at a cost of $32,899, or $146 per ha. Both of the sprinkler systems would
be inefficient in reducing nitrogen losses.
The above comparisons are valid only under the conditions assumed for
this model. Increases in the value of fertilizer, water, or labor could
cause some shifts in comparative advantages among the various irrigation
systems. Also, changes to more sandy soils or steeper slopes would auto-
matically reduce the advantage of surface irrigation systems over sprinkler
systems. Under the conditions assumed, a fanner operating in the Yakima
Basin with full technical and economic knowledge and complete mobility of
resources would elect to irrigate with system b, the surface tailwater reuse
system, which maximizes his net revenue.
The model was also used to look at additional soil types and slope
conditions. Figure 32 summarizes the marginal abatement costs for estimated
nitrogen leaching on alternate soil types in the Yakima region. Costs of
abatement can quickly reach $4/kg of nitrogen loss abatement if nitrogen
108
-------
UJ
t— l
i—
Qi
UJ
1—
— 1
-
UJ
Z. _J
0 -1
1— >
CD <
C£ I-H
rv \/
"— ' O
i — i y_
c3^ OO
o: s:
*Si UJ
0- r—
2! 00
O >-
0 00
•
CTl
.—l
UJ
1
CO
c^
I—
SI
1—
00
00
-^
o
t— 1
I—
3
2
o;
t- 4->
i — t/>
C "O
•1— •!—
£_ , —
Q. O
00 00
OJ •—
i — O
\S If
C 1
•r- OJ
t- TD
Q.-1-
00 00
o
QJ •!-
O +-> O)
ro ro to
4- E 3
£_ 0 OJ
^ -I-J Di
00 3
c_
CD CU
O -^ CU
ro ro t/)
4- S 3
l_ r- QJ
OO ro
i
r~^
CU O
O ro
ro J3
4- 1
t— 1^
rs ^
00 O
^ ^
QJ
x *
^ — ^
TD
,^ — ^
O
^-^
^ — ^
JO
* — ^
^ — ^
ro
N — ^
cu
to
ro
CQ
^j- r~~ to CTI
O • CTl CTl
r~~ to o O •— i co
»CM •» -
tO <=)- CM
LO OO
CTl CO «d- •— 1
CM • 00 <-O
O 00 O O O C
*CM « "
^-1 CO CT.
1 — 1 1 —
O1 OO CD to
tO • tO CM
CTl "vl- O O O O
*CM « r.
to «d- oo
CO
OO CTl t— 1 to
to • O O
CTl OO O O CO f^-
«CM •.
CO r-.
CO
r^ oo co co co t^-
o • to o i— i r-~-
C~*l t—H tO ^^* tO ^J"
«CM « « "
O i— 1 OO OO
,-H CO
r^ oo «* co LO
oo «— i r- oo o o
f\ CM « n *»
oo t— i t— i r-
,-H CO
^^ ^-^
CD. tO ^~~
-^ E c E
^ — " D- O Q.
D_-(-> Q. -K
~o * ^^ — ^^ — ' • — ^
cu *~~^ to
_C . T3 •*'»•
0 O i— O - —
ro C QJ C +-> QJ
O) O -r- O QJ to to
_i o >-, o 3 o :r>
C C_)
C C 4-> +-> Ol O)
QJ CU C C > i— O
CD CD CU CU CU O £_
O O E E O^ 5- ^
C ^_ >| — if— ^_> Q
-!-> -I-J T3 T5 4-> C tO
•i- -i- CU *
tO i-H
OO
. »
en
-^
0
^D • — .
0 E
^H O ~
— i to
ro c_
C r- c-
CU — '---
CD
0 t_ S-
1- CU O
+->+-> JO
•i— ro ro
•Z. 3 _J
•
QJ
to
>}
to
C
0
•1 —
+J
ro
en
•1 —
£-
•r-
cu
-p
CT)
C
•1 —
01
C
ro
.C
o
o
QJ
^3
-o
QJ
3
C
O)
cu
t —
ro
•r~-
-l~>
•1 —
c:
•r—
c
•r—
c
o
•r—
-I-J
o
33
~C3
QJ
ce
-K
109
-------
leaching is to be reduced below 50 percent of current levels for the model
farms. These estimated abatement costs are based on most efficient combina-
tions of new irrigation methods, management inputs, and crops for each level
of abatement (see Appendix B).
Figure 32. Long-run marginal cost function for the abatement of
nitrogen leaching for various soil types in the
Yakima Valley.
co
t-
V)
O
o
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
Loamy Sand
Clay Loam
14 II 9 7521
NITROGEN RESIDUALS, 1000 kg
Developing an Optimal Control Strategy (Step 7)
The ten policy options were first imposed on the farm assuming short-
term planning, with investment in additional irrigation equipment not being
allowed. The policies were also examined assuming long-range planning where
irrigation system changes would be feasible.
Production costs were derived from cost studies conducted by the
Cooperative Extension Service of Washington State University. Costs for
irrigation labor, nitrogen fertilizer, and electric power were entered in
the programming model at rates of $3.00 per hour, $0.44 per kg, and $0.082
per ha-cm ($0.085 per acre inch) of water, respectively. All costs and
prices used for this analysis were representative of the mid-1970s.
110
-------
Sediment yield is a function of only one production input - water.
Thus, controlling water use is the most economically effective method of
controlling most sediment losses. Under any policy to induce sediment
abatement, such as taxing sediment, constraining sediment, constraining
water use, or taxing water, the farmer would probably make the following
adjustments: As allowable sediment loss levels were reduced, the farmer
would first convert his total acreage to system a, thereby reducing sediment
outflow by 60 percent. This adjustment would have a total cost of about
$3,000. To reduce sediment loss still further, the farmer would find it
most profitable to substitute close growing crops for row crops. A subsidy
to the farmer would probably be required to induce such substitutions among
crops. A sediment outflow from the farm of 363 tonnes (80 percent reduc-
tion) would cost about $17,000 for the combination of practices. Minimum
outflow from the farm for the conditions examined (26 tonnes) would be
obtained by eliminating row crops. If sediment were the primary focus of
abatement, it would be less costly in the long-run to turn to tailwater
reuse systems or sprinkler systems.
A policy of subsidizing the farmer for increasing acreage of close
growing crops at the expense of high value row crops would necessitate an
unreasonably large subsidy. Due to the greater returns from row crops as
compared to close growing crops in this case, the farmer would probably
not be willing to alter his crop mix until subsidies exceeded $346 per
hectare (current return for row crops). Even this level of subsidy would
not be particularly efficient in reducing sediment yield. As discussed in
Section 3, it is probably not practical to attempt crop pattern adjustments
through subsidy programs of this type.
WATERSHED MODEL
Determining Applicable Control Measures7(Step 3)
Applying NPS controls to the entire watershed enables consideration of
a wider range of control practices and an overall assessment of control
strategies to achieve water quality objectives. Water quality criteria of
concern in the Basin include a maximum nitrogen concentration of .3 mg/
liter, a maximum river water temperature of 70°F in August, and a maximum
sediment loss of 22 Mt/ha (1 ton per acre) throughout the basin. Where
improved irrigation efficiency was needed to solve water quality problems,
investments in new irrigation systems, improvement in irrigation management,
or changes in crop mix or location of crop production within the basin were
required. The options include:
7The watershed evaluation of nonpoint source controls is based on work by
Pfeiffer (1975).
Ill
-------
Tailwater Reuse (IA-27)8, Sprinkler Irrigation (IA-29) and Improved Water
Management (IA-12) --
This alternative would increase surface irrigation efficiency by 10%
due to improved water management. The increases in efficiency are largely
achieved by a 5% increase in tailwater reuse and sprinkler irrigation sys-
tems.
Tax on Nitrogen Fertilizer (IA-32) --
This policy was chosen because nitrogen loss is primarily a function of
nitrogen application rate. The varying tax ranged from $0.22 to $1.54 per
kg ($0.10 to $0.70 per pound) of nitrogen fertilizer.
Water Charge (1-33) —
This option imposed a charge on the use of irrigation water ranging
from $0.62 to $3.08 per ha-meter ($5 to $25 per acre foot). Use of water
for irrigation reduces river flow, which in turn increases both the effec-
tive concentration of pollutants and the river water temperature. Moreover,
nitrogen loss is functionally related to the volume of deep percolation
water, and sediment loss is functionally related to the volume of runoff
water.
Uniform Reduction of Water Rights (IA-34) --
This practice imposed a uniform percentage reduction of water rights to
all regions, with the reductions varying from 10 percent to 60 percent of
current levels. Sales or exchanges of water rights between regions were not
permitted. This policy would result in efficient resource allocation only
if the demand for water is the same in all regions before and after water
rights are reduced. The major advantage of reducing water rights is the
fact that the policy does not require a direct income transfer from agricul-
ture to the public sector.
Nitrogen Tax (IA-32) and Charge on Water (IA-33) --
This alternative combined a tax on nitrogen ranging from $0.44 to $0.88
per kg ($0.20 to $0.40 per Ib. ) with a charge on water ranging from $0.62 to
$2.46 per ha-meter ($5 to $20 per acre foot). Since both water and nitrogen
are contributors to pollution through irrigation return flows, it was ex-
pected that a combination of policies would be more efficient than policies
directed at either individually.
Nitrogen Tax (IA-32) and Reduction in Water Rights (IA-34) --
Like the above option, this alternative combined a tax on nitrogen
ranging from $0.44 to $0.88/kg ($0.20 to $0.40 per Ib.) but with a reduction
in water rights from 10 percent to 50 percent throughout the area.
8Refer to Table 3.2 for a description of practices.
112
-------
Change all Furrow Irrigation Systems to Either Sprinkler (IA-29) or Reuse
Systems (IA-27)
See discussions on Sprinkler and Reuse Systems above.
Choosing the Unit of Analysis (Step 4) --
The watershed model characterizes both the agricultural system and the
water quality of the entire Yakima River Basin. The river is divided into 7
reaches (Figure 31) with an associated land area corresponding to each river
reach. In most cases, return flows from the farming region drain into the
respective reach.
Establishing the Base Condition (Step 5) --
Table 20 shows two solutions that were used in the watershed model9 to
establish a basis for comparison of environmental improvement policies.
Solution B! (base condition) constrained the crop area in all regions to
existing levels. This solution most accurately reflects current conditions
in the basin. Solution B2 permitted regional crop production specializa-
tion, by allowing a 25 percent increase in row, fruit and vegetable crop
area, a 50 percent increase in field crop area, and an unlimited increase in
forage crop production for each region. Basinwide areas of each crop were
still constrained to existing levels, however. The results of solution 82
were used as a base with which to compare all remaining policy results.
Evaluating Control Measures (Step 6)
The cost effectiveness analysis of control practices utilized three
mathematical models - an agricultural submodel, an hydrology submodel and a
water quality simulation model. The agricultural model limits crop acre-
ages, pollutant losses and practice application rates. These implicit con-
straints lead to abatement cost estimates. The hydrology submodel is a
watershed mass balance of water flows including precipitation, runoff, and
irrigation diversions for each subwatershed. Finally the water quality
simulation model relates agricultural and irrigation practices on cropland
to gross pollutant loading. Figure 33 shows inputs arid outputs of each
model and how they are linked when used to make policy decisions.
The Agricultural Submodel --
The agricultural submodel for the river basin included 10 crops, 3
irrigation systems, 11 regions and subregions, and 4 levels of nitrogen
application for most crops. In order to permit some degree of regional
specialization, row crop and fruit crop acreages were permitted to increase
25 percent and 50 percent, respectively, within each region. The total area
of these crops within the basin, however, was constrained to present
levels. The area of forage crops was not specifically restricted. The
model reflected surface and subsurface return flows from crop activities
within each region.
'For details of the model and input data used see Pfeiffer (1976).
113
-------
=z
oo
03
a:
LU
1 — 1
rY*
y
i — i
o 01 oo =3- CM LT> r-~ <=}- o
r*~ CO ^v. ^3" OO O^ ^^ CD CO ^~ O""> OO
O^i 00 iQQ CO ^J~ C\J • • l"*™1^ • •
*» r— * c\j r^- o o oo ^d"
^O CNJ C\J
o <3-
r™
C? ^" CO ^O OO ^3 OO CD ^"
LO OO O^ ^~ 0s! CT) r— r^ O"^ r— CO r—
O^ 00 CO OO ^f CSJ vO . .
** r~~ ** CVl P^- fTi CD CO *sj*
LO CM CM
O 'd-
1 —
E S
r i
L4_>
CD re re ZEZ
re O) NX c~ c** i — r^ re
-E ^ re -~^ \ o -E
-E O O O CT> O5 O - —
OOO^^OOO EEO-t->U
OOOCDOOO "2!o
OOO^i — i — i — r —
re re
E i ^
0 tO
-P 00 3
co re cj> i —
rs CO =t -P S-
3 O 3
r^ ^C " " -P "P
CO re E E re
E "O ~P "O ** O O *^ S-
O QJ O r" • CU re *r— gr- CO (/) D
o -p -P re -p E -p ~P co co Q-
E re -p t- o re re o o E
•i-Cn " "OCU-i-S-S-i — i — CU
Q.S-CDCU T- EE-P-P
os- -r- -r- "•O-PCUOJEEE
S--I— r— i — +J reoogj
-------
Nitrogen loss functions, sediment loss functions, and irrigation
efficiency relationships (see Appendix B) for various irrigation systems are
examples of the data requirements for this type of analysis. Sediment
losses were adjusted for factors such as crop cover, tillage practices,
method and rate of irrigation, slope and soil type, and overall management
level. Water rights in the river basin were assumed to represent maximum
allowable diversions of water. In most cases, water quantities are more
than sufficient to supply crop needs under current irrigation practices. By
imposing constraints and/or higher prices on inputs used by the agricultural
sector, abatement effectiveness and the distribution of abatement costs
resulting from alternative policies could be estimated and compared.
Hydrology Submodel ~
The linear programming model developed included a hydrology submodel
representing river flows in the basin. This hydrology model reflected
effects of water demands and return flows from the agricultural submodel on
river flows. Irrigation diversions and deliveries at each reach terminus,
and irrigation return flows carrying pollutants to the river, were included
in the model. The net flow of water at each reach terminus was taken as the
sum of water inflows minus water outflows for that reach. Natural inflows
were included by individual river reach, and diversion rights were allocated
by canal for each irrigation system of the river basin. In cases where
canals serve more than one region, the activities were appropriately sub-
divided. Water flows in each major reach of the river were input for the
period April through September. August, the period of lowest flow and most
serious water quality problems, was used to meet the water quality stan-
dards.
Water Quality Simulation Model --
The water quality simulation model was linked with the linear pro-
gramming model as illustrated in Figure 33. This model measured, for
example, the effects of nitrate concentrations contributed by irrigation
return flows. It was demonstrated that not all nitrogen entering return
flow channels from individual farms arrives at the river. Some nitrogen is
lost through denitrification or by transformation to organic forms during
plant and phytoplankton consumption (see Section 2). Once in the river,
phytoplankton use of nitrogen continues. Part of the nitrogen is diverted
with the irrigation water for additional use on farm land, and part remains
in the river as inorganic nitrate nitrogen. Thus, the nitrogen concentra-
tion in the river depends upon nitrogen losses from upstream farms, denitri-
fication, plant and phytoplankton uptake, subsequent irrigation diversions,
and the volume of water in the river itself.
The simulation model measured the flow of water and nitrogen throughout
the various reaches of the river, adding and subtracting nitrogen and water
as they were discharged into and diverted from the river. A pollutant
delivery ratio (PDR) of .32 was estimated, reflecting the nitrogen lost from
farms which actually enters the river in nitrate form during the irrigation
season. During August, the PDR is reduced further to approximately .24,
because of plant and phytoplankton uptake.
115
-------
CO
re
CO
O)
re
re
>-
OJ
co
O
+J
c:
o
o
ai
re
E
O)
re
or.
<
UJ
z
_i
^
x. *
XT
O
P*
g§
O ^™
z o
X"
>X
1
0
z
V
5
^1 tf
1
^
)-
J ;
< <
3 -
Qf 1
(E .
uj :
^ I
*«
j
z
3
j
JJT
2^
!c
02
V
S
i
—f '
xt
X
-
ij
3 (O
LJO
0) -J
X
V
1
•v
s
s.
X
t
"k
0
LU
„ ^™
ft*
~>
$ 0
s.
Js.
|X
X
a
u
x
c
X
>-
ts
o
J
o
or
o
1
.._ _ .
1
oc
^
5
Ei-
K
jvi
t
x^^
o
i
u.
Z UJ
33
UJ O
or >
X.
e>
z
_i ^
UJ ^
/V rf
5 < or
TUJ o
mzo
co^Q-
t
1
i
i
i
i
i
i
i
i
i
i
i
1
1
i
i
i
i
i
i
i
1
i
ixf :
^X, '
1
1
1
1
SI 1
E u. i
X^ i
1s — ,
r1
i
i
i
i
i
i
i
i
i
i
i
j i
i
116
-------
Results --
Table 21 summarizes the policies evaluated which met the prescribed
water quality standards. The table includes base solution 2 (B2) for com-
parison. Two measures of abatement costs were used. "Producer cost" mea-
sured the reduction of farm income caused by abatement policies, reflecting
the immediate and direct impact on agriculture. "Net social cost" measured
the public or total cost of abatement policies. The two differed if a
transfer of income from agriculture to the public (a tax) or from the public
to agriculture (a subsidy) was involved. Comparing the social costs of
alternative policies is a more realistic measure of their relative economic
efficiencies (see Appendix F for a further discussion of this topic).
Tax on Nitrogen Fertilizer (IA-32) -- As expected, nitrogen application
rates declined as the tax increased. Fertilizer use was reduced on all
crops at the higher tax rates, but reductions on the irrigated pasture (a
relatively low-value crop) were most pronounced. A nitrogen tax above
$0.88/kg ($0.40 per Ib.) resulted in no nitrogen fertilization of pasture in
the basin. Water diversions declined with irrigated acreage, so less pro-
duction of low intensity pasture caused August river flow volume to
increase. Nitrogen concentration and water temperature goals were finally
met with a nitrogen tax of $1.32/kg ($0.60 per Ib.).
Water Charge IA-33) — Irrigated pasture was once again the crop most
affected, though alfalfa acreage also declined sharply at the $162/ha-m ($20
per acrefoot) rate. The area of other crops remained relatively unchanged,
despite changes to more efficient irrigation methods.
Uniform Reduction of Water Rights (IA-34) -- Under this policy, as would be
expected, irrigated acreage declined rapidly as water rights were reduced,
primarily affecting crops in the order of their value. Water conserving
irrigation systems were also implemented with increasing frequency as water
rights were reduced. Water rights reductions of 30 percent or more even
caused fruit production to decline in some areas.
Prohibiting Surface Runoff (IA-29, IA-27) -- The result was almost no change
in cropping pattern and only a small reduction of income compared with the
base solution. Sediment loss was completely eliminated, but river flow
volume, nitrogen concentration and water temperature changed little. This
solution illustrated a policy directed at a single effluent, sediment,with-
out affecting other major pollutants in the river basin. This demonstrates
that it is possible to completely abate one pollutant in such a river basin
without significantly affecting other water quality parameters. The major
shortcoming of this policy was the fact that it had almost no impact on
either nitrogen concentration or water temperature, which are also serious
problems in the basin.
Developing an Optimal Control Strategy (Step 7)
Optimal abatement is possible either with taxes or with constraints on
inputs which contribute to a particular pollutant. Nitrogen content of
return flows, for example, would be most efficiently controlled with a
117
-------
B
o
i
coooo
8
it— en ro tn i —
< — CTiCM^OtJDCM
I — o»— cr»cr>
oroLO
IVOr^CTt*— ^•COCM*C> —
iOrotDtncnCf>CMCr>co
l< — LOCTiCJvCOCQ' — IDi —
1— CO
m c\j
o
O*3-CMCO'!j-OLO
vo
o
I— Q.
u_ LU
O I—
f
01 *o
§ s
g
O CJ1
SO
C7>-*-»COl
Ot-UUfOOOU-r-'r- ""U
3fO(/1 (- S- O<— <— +->
at a» -M-4-»i_ O O-M-M-M
(O ro S- S_ (D OJ OJ
ro O O -O O CU E
l_ r- •— O) r- »— Ol
•*-> fO +-> JD -4->
C C C-*™*^ 4-»(O+J'O
a>o cuvi cr— coiccnE
118
-------
policy affecting both nitrogen fertilizer and irrigation water use. Water
temperature, on the other hand, is a function primarily of river flow
volume, thus policies reducing water diversions would be most efficient in
affecting this quality parameter. Sediment loss, a function of surface
water runoff, would be most efficiently controlled by irrigation systems
which reduce or eliminate runoff volumes.
Table 22 shows that improved management (IA-27, IA-29 and IA-12) was
unable to meet the water quality standards which were sought for this river
basin. Each of the other solutions, d through h, was able to meet water
quality standards at varying levels of efficiency. In all cases, the net
cost for meeting water quality standards in the river basin were rather
large. The most efficient policy was solution g. It included a combination
of water charge and nitrogen tax in order to meet desired water quality
standards at a net cost of $9 million for the river basin. The least effi-
cient policy in meeting all quality standards was solution d, which taxed
only nitrogen. This policy resulted in a net social cost of over $1 mil-
lion. Net farm income would actually be decreased by even larger amounts
in each case, due to the additional charges to farmers for water and fer-
tilizer.
TABLE 22. INCOME, RESOURCE USE, AND WATER QUALITY IMPACTS ASSOCIATED WITH IMPROVED
MANAGEMENT (IA-27, IA-29 + IA-12)
Item Unit Base Solution B2 IA-27 + IA-29+IA-12b
Net crop income
Irrigated area
N applied, total
N applied per unit area
N lost, total
Water diverted
Flow at Kiona, August
N concentration, season
N concentration, August
Sediment loss. total
Sediment per irrigated
acre
Maximum temperature
$1 ,000
1,000 ha
1,000 kg
kg/ha
1,000 kg
1,000 ha-m
1 ,000 ha-m
mg/1
mg/1
1,000 mt
mt o
C
106,910
183
42,379
234
1,433
294
12
0.65
0.87
714
1.58
24.2
108,332
183
42,907
234
6,454
271
14
0.54
0.69
341
.75
23.3
a Maximum concentration.
b See Table 3 for description of control alternatives and practices.
The net social cost figures include required subsidies that must be
paid to agriuculture in order to meet specified water quality standards. To
induce similar changes in cropping patterns, nitrogen fertilization levels,
119
-------
or irrigation systems through subsidy programs would require annual costs
equal to net social costs. It is doubtful that benefits achieved through
water quality improvements to such levels would equal the costs imposed.
Although it had the highest social cost, solution f had the lowest
producer cost. It was some $8.6 million less expensive to the producers
than the next least expensive policy, solution h. The fact that no taxes or
charges were levied accounts for the low costs to producers imposed by solu-
tion f.
If subsidies could not be used to meet environmental standards, a
reduction of water rights (IA-34) would probably be considered the next most
acceptable solution by the farm sector, though the capital investment
required for improved irrigation systems would be large. The only policy
not requiring extensive capital investment for irrigation systems would be
solution d, although this policy had the highest producer costs of any
policies evaluated. It would not likely be a politically acceptable way of
meeting the proposed environmental standards.
Because of their relatively low value, the production of irrigated
pasture, alfalfa, and corn was the first to be reduced by all abatement
policies. The livestock industry, which was not included in the analysis,
would decline markedly as feed production declined under such policies.
Farm income would also be affected to a larger degree than indicated by the
model results, because livestock was not included in the analysis.
Labor employment in the farm, agricultural supply, and agricultural
processing sectors would decline under all policies analyzed. In addition
to farm output, the adoption of improved irrigation systems would lower
irrigation labor demand. Reduced farm income and its impact on the economy
would also diminish non-farm employment opportunities.
The watershed model showed that it is possible to improve water quality
in such a river basin by controlling agricultural inputs or activities.
Improvement to a point where water quality is suitable for virtually all
uses of the Yakima River required a reduction of farm income ranging from 16
to 41 percent, depending on the policy employed. In addition to reduced
farm income, these policies would impose burdens on the agricultural sector
in the form of reduced land values, diminished livestock numbers, decreased
activity employment, and increased capital investments in new irrigation
systems.
It should be noted that the costs of abatement in this analysis are
valid only as relative values among alternative policies. The real cost of
any policy might vary from that presented here because of changes in crop
input cost, crop prices, administrative program cost, and level of abatement
desired. (See Section 3.)
The historic pattern of water use in the Yakima River Basin has
promoted irrigation methods and practices which are not water conserving.
The policy of providing water at a fixed cost per unit area encourages the
substitution of water for management, labor, and other farm inputs. This
120
-------
has led to unnecessarily high amounts of surface runoff and deep percolation
loss, depleting the river flow and resulting in water quality degradation.
Under these circumstances, changes in water use policies would greatly
improve irrigation efficiency and river water quality.
121
-------
REFERENCES
Bailey, G.W., A.P. Barnett, W.R. Payne, and C.N. Smith. 1974. Herbicide
Runoff from Farm Coastal Plain Soil Types. EPA-660/2-74-017. U.S.
Environmental Protection Agency, Washington, D.C.
Bliven, L., e_t al_. Statistical Sampling to Evaluate Rural Water Quality.
Jour. Environ. Eng. Div., Proc. Amer. Soc. Civil Engr. In Review.
Brady, N.C. 1974. The Nature and Properties of Soils (8th ed.). McMillan,
New York.
Carter, H.O. and K.D. Cocks. 1968. Microgoal Functions and Economic Plann-
ing. Amer. J. Agr. Econ. pp. 400-411.
Chapra, S.C. 1980. Application of the Phosphorus Loading Concept to the
Great Lakes. In: Phosphorus Management Strategies for Lakes, R.C.
Loehr, C.S. Martin, W. Rast (eds.). Ann Arbor Science, Ann Arbor,
Michigan, pp. 135-152.
CNI. 1967. Conservation Needs Inventory. U.S.D.A./Soil Conservation
Service, Washington, D.C.
Davidson, J.M., G.H. Brusewitz, D.R. Baker and A.L. Wood. 1975. Use of
Soil Parameters for Describing Pesticide Movement through Soils. EPA
$00/2-75-009. U.S. Environmental Protection Agency, Corvallis, Oregon.
Ditoro, D.M. 1980. The Effect of Phosphorus Loading on Dissolved Oxygen in
Lake Erie. In: Phosphorus Management Strategies for Lakes, R.C.
Loehr, C.S. Martin, W. Rast (eds.). Ann Arbor Science, Ann Arbor,
Michigan, pp. 191-206.
EPA Guidelines. 1976. Guidelines for State and Areawide Water Quality
Management Program Development. U.S. Environmental Protection Agency.
Washington, D.C.
Gossett, D.L. 1975. The Economics of Changing the Water Quality of Irriga-
tion Return Flow from Farms in Central Washington. M.S. Thesis.
Washington State University.
Haith, D.A. and L.J. Tubbs. 1978. Modeling Nutrient Export in Rainfall and
Snowmelt Runoff. In: Best Management Practices for Agriculture and
Silviculture, R.C.Tbehr, D.A. Haith, M.F. Walter, C.S. Martin,
(eds.). Ann Arbor Science, Ann Arbor, Michigan.
122
-------
Honeycreek Report (Draft). 1979. U.S. Army Corps of Engineers, Buffalo
District.
Law, J.P. Jr., and H. Bernard. 1975. Impact of Agricultural Pollutants in
Water Uses. Trans, of ASAE. 1354:474-478.
Nicol, K.J., E.O. Heady and W.C. Madsen. 1974. Models of Soil Loss, Land
and Water Use, Spatial Agricultural Structure, and the Environment.
Card Report 49T, The Center for Agricultural and Rural Development,
Iowa State Univ., Ames, Iowa.
Pfeiffer, G.H. 1976. Economic Impacts of Controlling Water Quality in an
Irrigated River Basin. Ph.D. Thesis. Washington State University.
Stem, 6. 1978. USLE Parameters. Soil Conservation Service. Medina, Ohio.
Thomann, R.V. and J.E. Segna. 1980. Dynamic Phytoplankton - Phosphorus
Model of Lake Ontario, pp. 153-190.
Triplett, G.B., D.M. VanDoren, and S.W. Bone. Dec. 1973. An Evaluation of
Ohio Soils in Relation to No-Tillage Corn Production. Ohio Agricul-
tural Research and Development Center. Bulletin 1068.
Venice Township survey, Seneca County, Ohio. Nov. 1976. A Summary of
Economic Data from the Agricultural Practices Survey by G. Becker and
D.L. Foster. Dept. of Agric. Econ., Ohio State University.
Von Rumker, R. et al_. 1975. Production, Distribution, Use and Environ-
mental Impact Potential of Selected Pesticides. EPA 540/1-74-001.
Water Quality Data for Material Transport Stations. 1978. Lake Erie Waste
Management Study. Technical Report Series, U.S. Army Corps of
Engineers. Buffalo, New York.
Wineman, J.J., W. Walker, J. Kuhner, D.V. Smith, P. Ginberg and S.J.
Robinson. 1979. Evaluation of Controls for Agricultural Nonpoint
Source Pollution. In: Best Management Practices for Agriculture and
Silviculture, R.D. Uoehr, D.A. Haith, M.F. Walter and C.S. Martin
(eds.). Ann Arbor Science, Ann Arbor, Michigan, pp. 599-624.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting Rainfall Erosion
Losses - A Guide to Conservation Planning. USDA, Handbook No. 537,
Washington, D.C.
123
-------
APPENDIX A
ESTIMATING NONPOINT SOURCE SEDIMENT AND NUTRIENT LOADINGS
FROM NON-IRRIGATED CROPLANDS
L.J. Tubbs and D.A. Haith
SECTION 1
INTRODUCTION
The selection of best management practices (BMPs) for agricultural
systems should be based on considerations of the economic, institutional,
and water quality impacts of each management scheme. This report provides a
means for estimating the amounts of crop nutrients which may enter surface
and groundwater systems from various cropping situations in the eastern and
central United States. These estimates may be used directly to predict
average, annual nutrient loadings from an existing or proposed cropping
system, or they may be combined with economic models to determine the
distribution of management practices which efficiently meets water quality
objectives. Although the nutrient loading estimates provided in this report
are for individual fields under a single crop and cropping practice, a farm
or watershed system may be analyzed by summing the loading estimates for
each field in the system.
Several mechanisms for nutrient loss from cropland into receiving
waters are described in Figure A-l. These mechanisms may be divided into
field processes, transport or delivery processes, and stream processes.
Nutrient simulation models are often divided along these lines, resulting in
separate predictions of edge-of-field loadings, nutrient and sediment
transport, and surface water impacts. This report is concerned only with
the first two processes, which will be considered separately. Referring to
Figure A-l, direct runoff (surface runoff plus interflow or subsurface
runoff) may contain dissolved forms of nitrogen (N) and phosphorus (P), and
provides a means for transport of solid-phase N and P in eroded soil.
Percolated water may carry dissolved nutrients below the root zone to
underground aquifers. Dissolved N in groundwater may later reappear with
baseflow in streams. Dissolved P in deep percolation does not usually reach
surface waters due to the adsorption of phosphorus by soil particles below
the root zone.
The most general method for estimating nonpoint source nutrient losses
is through the use of mathematical simulation models. However, such models
124
-------
can be difficult to apply since they require a great deal of data and
intimate knowledge of the soil and crop environment being modelled.
Furthermore, available simulation models typically have high start-up costs
associated with specialized computer programs. In order to capture the
flexibility of simulation within an approach that can be applied relatively
easily by a water quality planner, the Cornell Nutrient Simulation (CNS)
model was run for ten-year periods using cropping and meteorologic data from
27 climatic areas in the eastern and central U.S. (Figure A-2). Within each
area, variations in soil water and fertility parameters were made, and
the results obtained were average annual water and dissolved nutrient losses
to edge-of-field. To facilitate their usefulness, the simulation results
were summarized as predictive equations for runoff, percolation, and
nutrient concentrations. Independent variables in the equations are curve
number (as in the U.S. Soil Conservation Service's runoff equation),
fertilizer applications, and soil nutrient levels. These predictive
equations are analogous to the Universal Soil Loss Equation (USLE), although
they are based on simulated rather than actual field experiments.
Although comparable simulation procedures could have been used to
derive general predictive equations for solid-phase nutrient losses, simpler
methods are available for estimating average annual losses. Since the USLE
provides estimates of average erosion rates, these soil losses can be
multiplied by N and P concentrations in sediment to determine annual solid-
phase nutrient losses. Concentrations in sediment are equal to in situ soil
concentrations multiplied by enrichment ratios to account for the selective
erosion of small organic matter and clay particles which have higher
nutrient contents than other soil constituents. This procedure for
estimating solid-phase nutrient losses assumes that levels of soil N and P
do not change significantly from year to year.
The remainder of this report consists of three sections and several
appendices. Section 2 documents the simulation experiments and their
results which were used to derive location-specific predictive equations for
runoff, percolation, and dissolved nutrient concentrations. Section 3
combines these results with the USLE to provide a general methodology for
estimating nonpoint source nutrient losses and evaluating potential best
management practices for their control. The methodology is demonstrated in
the final chapter by analyses of selected sample fields as well as an actual
watershed (Honey Creek, Ohio). The appendices describe the CNS model and
its application to the 27 climatic areas.
The methodology in this report presumes familiarity with, and access
to, two primary reference sources:
1. Stewart, B.A., Woolhiser, D.A., Wischmeier, W.H., Caro, J.H., and
M.H. Frere, Control of Water Pollution from Cropland - Vol. I and
_H, Report No. EPA-600/2-75-026a (Vol I), EPA-600/2-75-026b (Vol
II), U.S. Environmental Protection Agency, Washington, D.C., 1975,
1976.
2. Wischmeier, W.H. and D.D. Smith, Predicting Rainfall Erosion Losses
- A Guide to Conservation Planning, Agriculture Handbook No. 537,
U.S. Dept. of Agriculture, Washington, D.C., 1978.
125
-------
o
5
<
UJ
CO
uj
w
co
H- O
co Q:
CL
a:
O
a.
co
LL)
CO
CO
O
< O
cr cc
UJ
co
UJ
CO
CO
UJ
o
o
a:
Q.
to
l/l
o
O
O-
oo
E
co
.
O
cu
OJ
S-
126
-------
Figure A-2. Climatic Regions for Model Runs.
127
-------
SECTION 2
USE OF THE CNS MODEL TO DEVELOP PREDICTIVE EQUATIONS
FOR WATER LOSSES AND NUTRIENT CONCENTRATIONS
CNS MODEL DESCRIPTION
The CNS model computes water and nutrient balances for two adjacent
soil layers at depth 0-10 cm and 10-30 cm (Figure A-3). Soil water contents
of the two layers are updated daily, with any excess water over field
capacity in each layer percolating below that layer within one day. Direct
runoff on each day is predicted through the use of a modified form of the
U.S. Soil Conservation Service's curve number equation (Mockus, 1972), and
is a function of soil water content, rainfall plus snowmelt, and crop curve
number (CN). Solid-phase and dissolved N and P balances are maintained for
the upper soil layer and a dissolved N balance is kept for the lower soil
layer. Nutrient levels are updated monthly, and losses in runoff and
percolation are computed as functions of the average soil nutrient contents
during the month and total monthly runoff and percolation, as obtained from
the daily soil water model. The model operation is summarized in Figure A-4,
and the model equations are listed in Appendix I.
The CNS model presented a number of advantages for use in this study:
1) the model may be adjusted to simulate any of the major crops in the
eastern and central U.S.; 2) it does not require calibration with water
quality data; 3) low cost runs are possible for simulation periods suf-
ficiently long to estimate long-term average losses; and 4) model parameters
may be approximated using readily available secondary data sources and/or by
limited sampling in a study area. The primary disadvantages of the model
are: 1) some nutrient loss mechanisms (particularly ammonification and
dentrification) are not included; 2) the model is valid only for soils not
limited in drainage by a fragipan or otherwise impermeable layer; and 3) the
model has not been extensively validated, although validation studies have
been performed for manure-spread corn in New York and for corn in Georgia
(Appendix I).
The limitations of the CNS model also indicate limitations in the
nutrient loss predictive equations derived from the model results.
Specifically, the predictive equations are only applicable to soils that are
unrestricted in drainage, and ammonification and denitrification losses must
be subtracted from fertilizer or manure nitrogen inputs before applying the
predictive equations. The final problem is most relevant if the equations
are used to generate exact loading estimates from an existing or proposed
cropping scheme. In this case, field monitoring or water quality sampling
may be necessary in order to support the accuracy of the predictive
128
-------
PRECIPITATION
RAIN
V
SNOW
EVAPORATION AND
EVAPOTRANSPIRATION
4 4 DIRECT RUNOFF
MELT
(soil water)
SHALLOW PERCOLATION
( soil water)
DEEP PERCOLATION
10 cm
20 cm
FERTILIZER PRECIPITATION
INORGANIC N INORGANIC N
MANURE
INORGANIC N
(organic n)
(inorganic n)
MANURE
ORGANIC N
INORGANIC N IN
SHALLOW PERCOLATION
(inorganic n)
v
INORGANIC N IN
DEEP PERCOLATION
CROP N
UPTAKE
/
INORGANIC N
IN RUNOFF
10 cm
20 cm
Figure A-3. Soil water and nutrient balances in the CNS model.
129
-------
FERTILIZER
AVAILABLE P
MANURE
AVAILABLE P
V1 >'
(available p)
CROP P
UPTAKE
DISSOLVED P
IN RUNOFF
DISSOLVED P IN
SHALLOW PERCOLATION
10 cm
20 Cm
Figure A-3. (Continued)
130
-------
DAILY SOIL WATER MODEL
MONTHLY NUTRIENT MODEL
Inputs
Daily precipitation (cm)
Daily average air temperature
Crop development dates - (Julian)
Planting
Emergence
Full canopy attainment
Harvest
Soil available water capacity(cm/cm)
Crop curve numbers
Intermediate Outputs
Daily soil water levels (cm)
Evaporation and evapotrans-
piration (cm)
Snow accumulation and melt (cm)
Daily canopy development
Final Model Outputs
Daily direct runoff (cm)
Daily percolation (cm)
Inputs
Monthly runoff and percolation (cm)
Fertilizer and manure N and P
(kg/ha-yr)
Soil organic N and available P
(kg/ha-10 cm)
Crop development dates - (Julian)
Emergence
Maturity
Yearly crop N and P uptake
(kg/ha-yr)
o
Soil bulk density (g/cm )
Soil clay content(%) and pH
Yearly mineralization rate (%)
Intermediate Outputs
Average monthly soil levels of
dissolved N and dissolved P
(kg/ha-10 cm)
Monthly crop N and P uptakes
(kg/ha)
Monthly organic N mineraliza-
tion (kg/ha-10 cm)
Final Model Outputs
Dissolved N in runoff (kg/ha-mo)
Dissolved P in runoff (kg/ha-mo)
Dissolved N in percolation
(kg/ha-mo)
Figure A-4. Flowchart of the CNS model operation.
131
-------
equations. However, if the equations are to be used to judge the relative
effectiveness of alternative measures for reducing pollutant losses,
particularly in an economic analysis, this sampling may not be needed.
CNS MODEL IMPLEMENTATION
The CNS model as presented in Appendix I may be utilized directly to
generate edge-of-field nutrient losses given the appropriate meteorologic,
crop practice, and soils data for a specific field. However, direct
implementation of the CNS model is time-consuming and requires a detailed
understanding of the model interactions. To overcome this difficulty, the
CNS model was run for a large number of cropping situations In the eastern
and central U.S., and the results generalized into predictive equations for
edge-of-field loadings as a function of a subset of the model inputs. In
this way, the operation of the CNS model in converting the input parameters
into model results is replaced by the predictive equations (Figure A-5).
In order to apply the model over a large area while retaining the
sensitivity of the model to local crop and weather characteristics,
randomized meteorologic records and generalized field practice data was
substituted for specific field data inputs. Daily values for precipitation
and average temperature were generated stochastically using a first-order
Markov process (see Appendix II) fitted to the meteorologic characteristics
of the 27 geographic regions in Figure A-2. These regions were selected for
both consistent meteorologic characteristics (amount and timing of precipita-
tion and average temperature) and agricultural practices within each region.
The simulation period for each run of the model was selected to assure
results which approximate the long-term average losses that would be
predicted by an infinitely long simulation. It was found that ten-year
simulations generated average nutrient loss levels that fell consistently
within _+ 5% of those generated by 25- or 50-year runs. Therefore, the
simulation period was set at ten years for all crops in each region.
Soil organic N and available P levels were reinitialized to the same
fixed values at the beginning of each model year, whereas initial inorganic N
levels in both soil zones were set equal to the ending levels for the
previous year (i.e., inorganic N levels are not reinitialized). However, the
carryover of inorganic N from year to year was relatively small. The
simulation results aproximate steady-state cropping systems and do not
reflect long-term buildup or depletion of soil nutrients.
DERIVATION OF THE NUTRIENT LOADING PREDICTIVE EQUATIONS
Each input parameter to the CNS model falls in one of three categories,
according to its effects on the model outputs: 1) parameters which are fixed
or whose variations have a minimal effect on predicted nutrient loadings; 2)
parameters which interact according to a linear or other easily defined
function with model outputs; and 3) parameters which affect loading estimates
in a non-uniform (but significant) fashion. For example, the concentration
of dissolved N in precipitation is fixed for any area, while fertilization
rates vary substantially from one cropping system to another, and directly
affect estimated dissolved N or P loss in runoff and percolation.
132
-------
Figure A-5. Relations of the CNS model to the predictive equations.
fixed
inputs
variable inputs
\
\
individual field
characteristics
field-specific
parameter values
predictive
equations
predicted nutrient
loadings
field-specific
nutrient loadings
133
-------
Table A-l classifies the various model inputs according to this scheme,
for any crop and geographic region. The fixed parameters are not included in
the predictive equations, because they are assumed not to vary significantly
between different management options for a specific crop. The nonuniform
parameters cannot be related mathematically to the model results, but have a
substantial effect on the levels of nutrient losses. Consequently,
predictive equations must be derived for a range of each of these nonuniform
parameters in each geographic region in order to capture their effects on
edge-of-field loadings. This is reflected in the organization of Appendix
III, where, for each crop and region, equation coefficients are presented for
all combinations of tillage dates (spring or fall), fertilizer application
dates (spring or fall), soil hydrologic group (A, 8, C, or 0), and a range of
soil available water capacities (.05, .10, .15, and .20 cm of water/cm of
soil).
The remaining input parameters may be related directly to the model
results for any chosen values of the fixed and nonuniform inputs. These
relationships (the predictive equations) were designed for simplicity in
application and for sensitivity to variations in local field characteristics.
Average annual runoff and percolation as a percentage of yearly precipitation
is related to crop curve number by:
%R = a CN + b
%P = a' CN + b'
(1)
(2)
where %R
%P
average percent of annual precipitation appearing as direct runoff
average percent of annual precipitation which percolates below 30
cm.
CN =
m —
where
= crop curve number for average soil
moisture (antecedent moisture condition II)
a, b, a', b' = constants for each crop, soil, and cropping practice
Runoff and percolation percentages may be multiplied by the average annual
precipitation at a specific location to obtain average annual runoff and
percolation,
m
P -
(Pr)
where R = average annual direct runoff (cm)
iP_ = average annual percolation (cm)
Pr = average annual precipitation (cm)
134
-------
TABLE A-l. CLASSIFICATION OF CNS MODEL INPUTS ACCORDING TO THEIR
INFLUENCE ON PREDICTED NUTRIENT LOADINGS
FIXED INPUT PARAMETERS:
Meteorologic data
Dates of crop development
Crop N and P uptakes
Mineralization rates
Dissolved N concentration in precipitation
PARAMETERS WHICH VARY MODEL OUTPUTS CONTINUOUSLY:
Crop curve number
Fertilization rates of N and P
Soil organic N and available P contents
Soil P adsorption capacity
Soil bulk density
PARAMETERS WHICH VARY MODEL OUTPUTS NON-UNIFORMLY:
Soil available water capacity
Soil hydrologic group
Timing of fertilizer application
Timing of tillage operations
135
-------
?
The coefficient of determination (r ) between %R and CN and between %P and CN
was greater than 0.99 for all modelled crops and regions.
Similarly, the following predictive equations approximate the relation-
ship between the CNS model input parameters and average annual dissolved N
concentrations in runoff and percolation and average annual dissolved P
concentration in runoff.
KRN = a0 + ajCN + a2FN + a3SN + a4FNCN + a5SNCN (5)
KPN - bg + biCN + b2F|\| + b3S|\| + b4FNCN + b^CH (6)
A >S A
KRP = YP(CO + qCN + c2Fp + C3Sp + c4FpCN + csSpCN) (7)
where KR|\| = average concentration of dissolved N in runoff
= average concentration of dissolved N in percolation (mg/Jl)
= average concentration of dissolved P in runoff (yg/Ji)
r\
FN = annual fertilizer and manure inorganic N input (10 kg/ha-yr)
S|\j = soil organic N in the surface 10 cm (103 kg/ha)
Yp = adjustment factor for soil P adsorption capacity (dmless)
Fp = annual fertilizer and manure available P input (kg/ha-yr)
Sp = soil available P in the surface 10 cm (kg/ha)
a-j , b-j , c-j = constants for each crop, soil, and cropping practice
The choice of units for FN, Fp, S|\j, and Sp are such that the constants
a-j, b-j, and c-j may have the same approximate order of magnitude, for ease in
presentation in Appendix III. The coefficients of determination (r2) between
KRN and the right-hand terms of equation 5 and between Kp^ and the right-hand
terms of equation 6 are greater than 0.97 for all crops and regions modelled,
and the multiple correlation coefficient for equation 7 is always greater
than 0.99.
The soil P adsorption capacity adjustment factor yp is a fraction of
soil bulk density p and the phosphorus adsorption coefficient e, which in
turn is a function of soil pH and clay content:
136
-------
g = 5.1 + 2.2(%C) + 26.4(pH - 6.0)2 (8)
where 3 = soil adsorption coefficient for available P ( •; ,„ )
mg/ X
%C = percent clay content (soil particles < .002 mm) in the surface 10
cm of soil
pH = average pH of surface 10 cm of soil
In generating values for the coefficients in equation 7, the CNS model
was provided with reference values for soil bulk density p of 1.3 g/cm , clay
content of 15%, and a pH of 7.0, yielding a value for 3 of 65 . The
my /
factor Yp captures the effect on dissolved P concentrations in runoff of
differences in p and 3 from these reference values:
v - (1.3) (65) _ 84.5
YP " p 3 ~P~T
Sl
Methods for determining values of CN, F^, Fp, S^, Sp, and yp and for select-
ing the appropriate equation coefficients from Appendix III based on field
characteristics are detailed in Section 3.
DATA USED IN THE CNS SIMULATIONS
This section details the data acquisition methods for all runs performed
with the CNS model. The appropriate methods for selecting values for each
input parameter to the CNS model are dependent on the relationship between
that input parameter and the model results. Referring back to Table A-l, the
procedures for data acquisition are different for the fixed, continuous, and
nonuniform input parameters.
Fixed Input Parameters
The fixed parameters in this application of the CNS model do not vary
between simulations of different management options, although there will be
differences in the parameter values among geographic regions.
Meteorologic Data--
The meteorologic models in Appendix II require input data for average
precipitation and number of days with precipitation by month, and average
summer and annual air temperatures. Data was obtained from a single source
(National Oceanic and Atmospheric Administration, 1974), which summarizes the
meteorologic characteristics of a number of weather stations in each of the
fifty states. For each geographic region in Figure A-2, one station was
selected to provide precipitation and temperature data for that region
(Tables A-2 and A-3).
137
-------
o
CD
LLJ
C£
^_
§§
i—
oo
-j_
CJ
LU
a:
O
u_
o
^
z:
Q
i — i
<
I —
i — i
a.
o
LU
a:
a.
^
-^
oo
>-
ct
a
LL*.
o
Qi
LU
^
^
—^
O
^
^ ^
g
•^
p^
1 — 1
I—
1 — t
Q-
i— (
CJ
. 1 1
a;
n
1 1 1
CD
fy^
LU
^^
^
C\J
1
LU
-_J
CO
-•^^
! —
i — =3-
rO
-l-> OO OO
O «3- cn
cn
Q • LO
o
«3-
'Z. • LO
i —
cn
O • LO
CM
0
oo • to
oo
0
<: • oo
CM
•-D • cn
LO
JC LO
•4-> -O -00
C 00 r—
o
s:
21 • CM
I—
<: • co
LO
1^
s: • cn
CM
00
u_ • to
0
CM
1
0
•(— •
-(-> Q
(O
+J OO
oo
S- >,
a> -4-> to
-C •!- >>
4-> CJ (O
rC T3
S v- =t*=
Q.
fO
c
o
cn "~
(U
a:
^•O
•
P"*"* ^o
^- o
I —
LO
• CO
1 —
CM
• !*•»•
CM
CM
• to
oo
^.
• cn
cn
• cn
to
cn
• o
LO i—
^_
• f—,
r--. i—
LO
• ,.
LO r—
CO
• 00
^J-
oo
• cn
CM
1^
• CO
LO
• o
•
Q
"^ ^L
fO
o
S_
fC
u_
CM
to
•
OO LO
t~~ oo
cn
• oo
CM i —
LO
• r-~
•=t r-
LO
• cn
LO
oo
• r—
l^r-
(^
• r—
cn i—
0
cn i—
00
• oo
O r—
^*
• 00
00 •—
o
• r—
IO i —
, —
• r—
*^t" i —
^f-
• O
CM i—
cn
• CM
CM i—
•
c
• p-
21 to
>^
•> (O
1 •*
3 =tfe
~1
Q
OO
, —
t
«* oo
tocn
,^
• to
•""
CO
• LO
CM
^
• 10
oo
,_J_
• r^^-
^
00
• o
CO r—
to
• o
co
• 1 — -
O r-
to
• o
CO r—
o
• CO
«^-
• cn
oo
cn
• to
*~
CO
• LO
1
.
a
•
00
1\
to
r—
i — tO
fO >,
-a
X
3 =tt=
0
•r-
oo
•*
LO
•
00 CM
Tl
, —
• cn
oo
to
• en
"^
cn
• r~^
•=*"
rv>
• CO
cn
«_,.
• cn
00
CM
• CM
CO •—
00
cn i—
oo
• CM
CO r—
cn
• cn
LO
i —
• CM
^" r—
00
• to
CM
j
• CO
oo
•
o
to
"^:
" to
0) >,
tO fC
to "O
0
t- "-tfc
CJ
to
LO
to
^ cn
^t-r^
o
•~
^
• to
"~
to
• LO
CM
^_
» «^J"
"*
^.
• cn
LO
,
• 00
to
•vf
• cn
p^,
00
• cn
to
CM
• CO
LO
OO
• to
CM
cn
• LO
o
0
• to
1
•
.a
a;
*
>-J
4-> rt3
fO "O
r—
Q. ^
•
z
to
LO
•
«d- to
to co
LO
• «^J-
•~
cn
• LO
CM
^f.
• to
•*
CM
• r^
to
,
• cn
CO
to
• 00
CO
f^
• 0
CT!5 r—
i —
CM
cn r—
1^^
• cn
LO
^J-
• r^.
oo
CM
• LO
CM
LO
• LO
1
•
to
c
to
f
IO tO
•r- >>
T3 to
i. -a
o
O =tfe
c
o
CJ
r^
cn
i^x ^j*
CO O
,
• psx.
^"
co
• r^*»
"^
cn
• r*^
to
_
• cn
CO
CM
• CO
cn
to
• cn
CO
1 —
. ^->
CM i —
i —
CM
• CM
0 i—
p^.
* r— ~
CO r—
cn
• 0
tO r—
LO
• I**-*
OO
CM
• f^
"*
fO
S
0
1 — 1
«>
c~
O tO
+J 2>y
cn fO
c -O
•r—
r— ^fe
i.
3
CO
00
cn
LO O
«=i- to
t^
• oo
"~
^
• 00
•"
t
• LO
LO
o
• LO
to
OO
• to
«3-
,
• 00
LO
5^-
• f-^
to
1 —
• 00
00
cn
• ''^
CM
cn
• «^-
1 —
^J-
• ^J~
1
(^
• ^~
^~
•
X
O)
I—
to
« >^
^. to
o -a
0
JO =tfe
r*t
3
_J
cn
CM
•
cn LO
LO to
CM
• ^j-
oo
co
• LO
CM
CM
• LO
1^-
00
• to
LO
^
• LO
OO
00
• LO
LO
CO
• to
to
0
• 00
r~
CO
• 1--.
LO
to
• ^J~
CM
co
. LO
CM
CM
• LO
CM
•
X
>
O) tO
C TD
^
« to
c -o
•r"
+J ^te
to
3
cC
•—
"
138
-------
•* •»
~o
O)
Z3
C
Ir~
-I-5
£1
O
(— *,
'*—-'*
CvJ
1
LU
^j
CQ
«3- to
0 00
t ^_ ,__
to
Q . CO
LO
O
•Z. ' CO
^
to
o • r^
cn
cc
t/i • co
cn
•=3"
eC • cn
CO
cn
•-3 • cn
cn
^ CO
•*-> ra • o
C CM r-
O r—
,3.
^* • r-~
OO i—
^.
,
+-> t|_ >
ro ea
-*: -o
S-
ro ~tfc
X
CU
f—
oo
CM
•
cn r — *
CM O
CM
• 0
oo
• CO
0
LO
• i_f")
LO
OO
» p^
LO
OO
• r^.
00
(^
• 1—
^ p~
«_J-
• cn
cn
^
• cn
0
CM
• cn
CM
CO
• 0
LO i —
r— •
cn
CM r—
cn
CM r^
•
to
to
^^
to
H ^j
c: ro
o -o
to
-M =tt=
O
ro
•"3
*
•
•— 0
to CM
to
• 0
CM
• CO
O
oo
. to
cn
CO
• o
•=3- •—
oo
• *^t"
to i—
o
• LO
CO •—
t
• oo
^3" r-^
•—
CO
• CT)
oo
CO
OO
^
• cn
to
to
• cn
o
r—
|
• O
CM t
ro
1
t\
to
c:
ro to
i— ro
O
S
o>
"ZL
LO
CM
•
cn r^^
r~- oo
CM
LO i—
oo
• *^}"
to i—
to
• cn
to
to
l^r-
cn
• i —
to i—
cn
• CO
to
^.
• fx^
00
00
• 00
CO
LO
• oo
r-^ •—
00
• to
LO 1 —
^
• CM
«3- r—
cn
• CM
pf- __
*^ f^^
r-
o
'£
t\
to
-o
•i—
Q. to
ro >,
•o
•o
ro
S-
CJ3
to
00
•
to cn
00 CM
• CM
LO i —
t
• o
to i—
^
• 00
to
CO
^
t
• oo
oo
• o
00 r-
o
• r-~
cn i—
^.
• CM
CO i—
oo
• oo
CO r—
o
• oo
00 r-
LO
. i—
*3- r-
LO
•o
c
1—4
*
O)
C CO
fO fO
Sg -Q
4-J ^tfe
^,
o
LJ_
^
o
•
i — OO
i— OO
00
• CM
cn r—
o
• o
00 i—
CM
• CO
to
CM
• CO
•^
to
• o
CO •—
00
• r-^
o •—
r—
^
• CM
O r-
o
• CM
cn r-
to
• CM
cn •—
oo
• oo
1— 1—
cn
• CM
CO •—
^
• oo
,
^.^
•*s
ft
c to
o >>
+-> rO
05 T3
C
><
^
to
^.
• CO
to
^
• ^—
CO i—
o
• CM
CM i—
oo
• CM
CO i—
,_
• CM
cn i—
CM
cn i—
oo
• oo
CM i—
o
• CM
CM i—
LO
• oo
s^
c—
CU
1 —
f\
cu to
i— ro
•i- "O
X =*=
o
E
^
cn
•
cn LO
•— •—
• o
^2 "~
LO
• 00
^
CM
• to
to
oo
• 1^
oo
!
• cn
cn
o
• CM
CM i—
f^
• 0
cn i—
0
• cn
00
•«*
• cn
to
• CM
OO r—
LO
• o
| .
OO
• "—
,
ro
CD
to
ro (O
•t-> T3
C
ro =fe
40
ct
O
CM
OO
•
to oo
OO i—
LO
• o
r— '
O
• p^.
o
o
• to
to
cn
• CO
CO
,
• r~~
CM r—
to
• CM
cn
r~— f—
•—
CO
• CTl
CO
CM
• CO
CM
LO
• O
IO i—
0
• O
CM i—
^
• •—
ro
e-^
^
>_
S- to
CU >5
o -a
Ol
•M =«*=
c
0
s:
CM
0
•
CM i—
cn to
LO
• CO
to •—
^j.
• *^J~
to i—
CO
• r—
r^ r—
LO
• cn
^
i —
• ^~
cn i—
^.
• o
cn •—
CO
• |
cn .—
cn
• oo
co •—
LO
• ^~
[^ r-
oo
• to
LO
• LO
LO r—
^.
• l~~-
>!
.
^y
».
C~
O tO
E to
to ~a
O1-:fr;
C
•1 —
ca
CM
CM
139
-------
*•""«*
-o
~i
c
•r—
j *
C
o
o
^_^
CNJ
1
<<
UJ
i
ca
ec
t-
i— CO
rd
+J CO LO
O O CM
CO
Q • r—
cn ,—
10
s: • CM
O r-
__
o • cn
CO
cn
00 • CO
CO
(
>
4-> C rO
rd ro "O
0) •—
3 4-> ^
^
O
Q.
C
O
•i- CO
CD CM
a>
a:
*
cn ro
CO CO
• CM
UD r—
cn
• CM
VO r-
o
• CO
t
• cn
cn
CO
• o
^^ n~
cn
CO ^
CO
• r—
CO r-
CO
• CO
CO i—
o
• CM
r^ r-
cn
• CM
tO r"*"
to
• ^~
LO I —
CO
• CO
tO i—
>_
z: to
^_
» ro
^} ^5
C
rO =fe
t~\
r—
>
O ro
E "O
• r—
-M =«=
i— ~
ra
ca
LO
CM
CO
•
O LO
CO i—
• cn
CO
00
f^-
to
lv-
o
• cn
to
^
• CM
1^ r—
i —
LO
• ^J-
cn •—
• —
CO
• CD
O •—
^~
cn
• cn
CO
ro
• CO
rv-
CM
• 0
O r—
^
• 0
CO r—
CM
• O
C_)
"^^
c
o to
*P ^)
CD ro
C ^D
tf_
£ ^ffr:
r*—
*r—
S
^O
CVJ
0
*
•— CO
co o
co
* ^O
•*
^
• LO
CO
t
• CO
^
LO
• CO
I^""
cn
• r^
O r—
CM
cn
• r~-
r^ r—
CM
LO
• CM
CO r-
r—
CM
• **o
rs"
CM
• LO
^
LO
• f^^
cn
CM
• r^
rs-
^
• to
LO
ro
I — tO
U_ >,
** "O
ro
Q_::te
E
ro
I—
I*"*",
C\J
140
-------
TABLE A-3. AVERAGE JULY AND AVERAGE ANNUAL TEMPERATURE (°C) FOR EACH
STUDY REGION
Region
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Average
July
22.4
21.8
18.6
23.8
23.3
24.3
26.7
24.8
26.4
28.4
29.3
26.0
28.1
27.8
28.6
22.4
23.1
24.8
25.8
26.1
27.3
20.2
20.1
22,3
24.9
26.7
27.8
Temperature
Annual
8.2
4.0
2.8
6.9
7.1
9.9
12.2
10.3
15.2
17.7
19.8
13.4
17.7
18.4
20.9
8.9
9.9
12.9
15.2
16.6
18.4
7.8
7.2
8.6
13.2
17.8
22.0
141
-------
Dates of Crop Development--
Normal ranges of the dates of planting and harvest by state for each
crop were obtained from published data (U.S. Department of Agriculture,
1972). The dates of planting and harvest for all years of the simulation in
each region were fixed at the mean dates for the states contained in each
region (Table A-4). Hay is grown continuously throughout the year, and
nutrient uptake was assumed to occur from 30 days before first cutting until
the last cutting in the fall. Dates of crop emergence, full canopy
attainment, and crop maturity were determined by adding a fixed crop develop-
ment period for each crop to the date of planting. For example, crop
emergence for corn in each region occurs approximately 14 days after planting
in that region, and full canopy is reached 75 days after planting. Further
crop development data is given in Table A-5. The variation in time to
maturity for corn by region is due to the use of shorter-season hybirds in
the northern U.S.
Crop Nutrient Uptakes--
The CNS model computes dissolved N and P uptakes by month as a function
of total yearly crop uptake and a crop development curve fitted to the dates
of crop development (see Appendix I). Average crop yields by region were
obtained from a recent agricultural census (U.S. Department of Agriculture,
1977) and were combined with crop nutrient contents (Martin and Leonard,
1967) to determine the yearly nutrient uptakes presented in Table A-6. For
corn, the yields for silage were used to determine crop uptakes. If corn is
harvested for grain, the N and P remaining in the stover will return to the
field in both organic and inorganic forms. Consequently, all N and P remain-
ing on the field in crop residues are to be added to the amount of these
nutrients applied in fertilizer or manure in each year. A substantial frac-
tion of the nitrogen needs of leguminous crops is met through fixation of
atmospheric N. It was assumed in the operation of the model that a legumi-
nous crop will utilize any available inorganic N before fixing atmospheric N.
Mineralization Rates —
The rate of nitrogen mineralization from organic to inorganic N is
roughly proportional to average temperature, and ranges from 2% to 4% per
year (Brady, 1974). Therefore, mineralization rates were chosen for each
region which reflect the average annual temperature of the station represent-
ing that region. The rates used in the CNS model runs are summarized in
Table A-7.
Dissolved N in Precipitation--
The amount of dissolved N in precipitation, although small, will affect
the concentrations of N loss in runoff and percolation for crops which are
fertilized at low levels. The total yearly loadings shown in Figure A-6
(McElroy, _et _al_., 1976) were divided by average annual precipitation in each
region to determine the average concentration of dissolved N in precipitation
in each region. The dissolved N input from precipitation in each month of
the simulation was then equal to the total precipitation for the month times
the average concentration for the region.
142
-------
o
CJ3
LU
O
^
O
•-3
a:
o
LL_
| —
LU
o:
^
O
^
Z
1 —
eC
GI-
LL.
0
oo
LU
1
LU
1
__J
ff^
h-
QJ
i > ^
-T— * ^
(C i-
Ol
00 C
D.
O
i"V +J
O CO
"O CD
C >
LU S-
ra
31
«*
:n
-|_>
C CO
•i- Ol
Cn >
01 S-
CQ n3
^
i^
QJ
1C
c:
S-
o
4J
fO
O
LU
V^^ |
CO 00
LO LO
fT\ ON
\J » W t
LO LO O
,—_ rm. ^\j
Of — i t — i
*^-J IM*
^O ^«O ^.D
0 0
CM CXI LO
'CTi CTi CTt
0 0
i — 1 — LO
^\ " — *^>»
tO IO U3
---,^.^.
LO
^~ r-~ C\j
r— CM CO
1 1 \ v~,^~-
r-r^r^
LO
CM CM ^^
LO O O O O
Of — \ ( — \ ( — \ f—\
k_^ l»_J \^^f \^_J
LO J — i— LO
I— r— "**+^ ^^^ n—
~-^^o o ^^
en cy> i — i — cr>
OOO
i — LO CM CM CM
"^^ ^^* "^^ ^^ ^^
CO U3 LO LO LO
O O O O LO
<^- LO 1^ CO
O O LO O O LO LO
10 lo^oioio r--i^
LO LO
O O O O O 1 1 O O
LO LO LO LO
000
LO
r— r— O
*^* " — CM
1 1 1 O I i I O ~-^
i — i — CTl
LO O
i — i — CM
1 1 1 -^^ 1 1 1 -"^ -^^
LO l£> LO
O 00
Of""} t — 1 f"~l f~"} t — 1 t — 1 r— ^~
**— J \~-J ^-/ ^__J (^ J \^J ^^ ^^
LOLOLO LOLOOLOLO
CTlOi — CMCO^-LOW3I —
LO LO O
f— . f/\j p |— f—ff f— _
f^s, V.Q ^Q oo r**^ r**^
OOO LO O
r^ C\J OJ CD C\J r—
CD CD CD 1 "^». 1 1 CD CD 1
LO LO LO LO LO
r— . ,__ f\j f— tO f~" r—
CD i — CD CD 1 1 1 n— r-— CD
o o
LO i LO i-~* C\J r-~ r—
o
LO i— i— O O 00 i—
^™* "*^s^ **^s. r~~ OJ "^-s^. **^^
^--x CD l 1 CD *"*»*. "*^x. CD CD I
o^ i — r— cr» cr^ i — i —
LO LO LO
r^ I — C\J LO I"- i — CM
"^V. "^^s. 1 1 *"*^N "***^ "^^ '"*fc^. """^ 1
LO LO LO VO (— \ —^ _^_ * — \ y— \ «.s.
t_j ^_j i^^ ^_j ^™ r^~ t^^ ^j ^^.
LO LO O
143
-------
o:
O-
O
LxJ
Q-
O
co
« t/0
UJ •
O ZD
03 • O
CSJ
un
co
o
o
o
o
t_3
LU
Q-
O
o;
o
i.
o
to
c
03
CD
.0
a>
c
o
O)
TD
E
n3
fC
CD
CM
IO
CTi
rC
CM
LO
r^
01
00
144
-------
TABLE A-6. AVERAGE ANNUAL NITROGEN (N) AND PHOSPHORUS (P) UPTAKES (KG/HA-YR)
FOR MAJOR CROPS IN EACH STUDY REGION
REGION
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Corn
N
38
38
68
68
83
75
75
90
112
112
105
75
83
90
83
83
105
112
105
90
83
98
98
105
105
98
83
P
9
9
16
16
20
18
18
22
27
27
25
18
20
22
20
20
25
27
25
22
20
23
23
25
25
23
20
Hay-
Alfalfa
N
80
80
150
150
160
150
150
165
-
-
-
-
-
-
-
145
160
-
-
-
-
150
125
145
-
-
-
CROP
P
7
7
12
12
13
12
12
14
-
-
-
-
-
-
-
12
13
-
-
-
-
12
11
12
-
-
_
Soybeans
N P
70
70
75
80
95
70
70
85
90
90
85
75
80
80
80
80
115
90
80
80
80
-
-
-
80
75
90
7
7
8
8
10
8
8
9
10
10
9
8
9
8
9
9
12
10
8
9
9
-
-
-
8
8
10
Winter
Wheat
N
38
38
-
-
-
38
38
40
25
25
28
40
34
-
-
48
50
40
40
32
-
45
-
-
40
38
-
P
8
8
-
-
-
8
8
8
5
5
6
8
7
-
-
10
10
8
8
6
-
9
-
-
8
8
-
145
-------
o
I—I
CD
LU
<
a:
ca
00
LU
O
o
Q
LU
oo
o
CD
C£
O
O
oo
C£L
O
LU
o:
LU
<
o;
Ln
un
o
LU
_
LU
o;
(NJ
tn
CM
o
oo
LU
o;
146
-------
Continuous Input Parameters
The parameters in this group may assume continuous values and will vary
model outputs in a continuous and well-defined fashion (the predictive
equations). For each combination of the fixed and nonuniform input para-
meters, the CNS model was replicated for a number of combinations of the
continuous inputs sufficient to determine the predictive equations. Because
the predictive equations varied uniformly with the independent variables, it
was necessary to repeat the model runs for each cropping situation for only
two values of curve number (CM), fertilization rate (F^, Fp), and soil
fertility level (S|\j, Sp) to determine the predictive equation coefficients.
Soil P adsorption capacity and bulk density were not varied between runs,
since their effects on dissolved P loss in runoff are inferred directly from
the model equations in Appendix I. The effect of variations in these factors
on dissolved P loadings is contained in the variable yp, as computed by
equation 9.
In summary, for each cropping alternative in each region, the CNS model
was run eight times, representing all combinations of two curve numbers, two
levels of fertilization, and two soil fertility levels. Curve numbers were
set at CNjj = 60 or CN jj= 90, soil organic N was initialized at 1000 kg/ha-
10 cm or 10,000 kg/ha-10 cm at the beginning of each model year, and soil
available P was initialized at 10 kg/ha or 200 kg/ha at the beginning of each
model year. Annual fertilizer N and P inputs were set at 50% or at 200% of
the normal fertilization rate for each crop. For corn, this range translates
to 50 kg/ha-yr or 200 kg/ha-yr of inorganic N in all regions, and 10 kg/ha-yr
or 70 kg/ha-yr of available P in all regions. Hay fertilization rates were
10 kg/ha-yr or 70 kg/ha-yr for both inorganic N and available P. Soybean
fertilization rates were 0 kg/ha-yr or 30 kg/ha-yr for incorganic N and 10
kg/ha-yr or 50 kg/ha-yr for available P.
Non-uniform Input Parameters
As mentioned previously, the non-uniform parameters have a substantial
effect on the levels of predicted nutrient loadings, but cannot be related
directly to the model results. Soil hydrologic group is a categorical
variable (A, B, C, or D), and determines the magnitude of direct runoff under
fallow conditions, independent of crop curve number. Soil water capacity,
although a continuous variable, exerts a nonuniform effect on nutrient loss.
Discrete values for soil water capacity of .05, .10, .15, and .20 cm of
water/cm of soil were modelled for all combinations of the continuous and
other nonuniform variables.
Timing of tillage and fertilizer application were both related to the
times of planting and harvest in each region. Fall plowing was assumed to
occur two weeks after crop harvest, and spring plowing occured one month
before planting. Fertilizer N and P were added to the nutrient budget in the
month after harvest for fall fertilization and in the month previous to
planting for spring fertilization. Hay was assumed to be fertilized in the
month previous to first cutting. No-tillage was modelled as continuous cover
(no bare-soil periods).
147
-------
148
-------
SUMMARY
The CMS model was used to estimate edge-of-field loadings of dissolved N
and P for a number of combinations of crops, soils, and field practices.
Loading estimates were condensed into predictive equations for water and
nutrient loss as a function of soil and crop characteristics. The coeffi-
cients for the predictive equations for each modelled crop are presented in
Appendix III. Table A-8 summarizes the various crops, soils, and field
practices modelled.
149
-------
TABLE A-8. CROPS AND PRACTICES SIMULATED WITH THE CNS MODEL
CROP
Corn
Hay
Soybeans
Corn with
Winter Wheat
TIME OF
PLOWING
Spring
Fall
Reduced Tillage
No Tillage
Spring
Fall
Reduced Tillage
No Tillage
TIME OF
FERTILIZATION
Spring
Fall
Spring
Fall
Spring
Fall
Spring
Spring
Fall
Spring
Fall
Spring
Fall
Spring
Fall
GEOGRAPHIC
REGIONS
all
all
all
all
all
all
1-8, 12, 16-19,
22-26
1-21, 25-27
1-21, 25-27
1-21, 25-27
1-21, 25-27
1-21, 25-27
1-21, 25-27
1, 2, 6-13, 16-20,
22, 25, 26
1, 2, 6-13, 16-20,
22, 25, 26
150
-------
SECTION 3
A GENERAL METHODOLOGY FOR ESTIMATING NONPOINT
SOURCE NUTRIENT LOADINGS
In this section, the predictive equations derived in Section 2 for
runoff, percolation, and dissolved nutrient loadings from cropland are
combined with the USLE to produce a methodology for predicting average annual
edge-of-field water and nutrient losses for several major field crops in the
eastern and central U.S. Methods are also presented for obtaining
approximate estimates of average nutrient delivery from edge-of-field to
surface waters.
Three objectives may be met through application of the methodology
presented in this section:
1. Dissolved and solid-phase pollutant loading may be estimated for a
specific field, given certain crop characteristics,
2. The effectiveness of a specific management practice in reducing or
eliminating pollutant loading as compared to base conditions (say,
the present cropping system) may be evaluated, and
3. Two or more control practices may be ranked according to their
effectiveness in reducing the loading of each pollutant.
Each of these objectives requires the computation of pollutant loading based
on particular field characteristics for each alternative practice. Table A-9
summarizes the required data, predictive equations, and the resulting pol-
lutant loading estimates for both dissolved and solid-phase pollutant
loadings.
Some field practices cannot be evaluated using the methodology presented
in this section. For example, changes in the average time of planting of a
crop may increase or decrease average pollutant loading; however, variations
in the time of planting or harvest are not included in the methodology. In
general, a practice may be evaluated if it can be represented by changes in
the input data parameters listed in Table A-9 for each pollutant. A number
of soil water and nutrient control practices are evaluated in Table A-10 for
their effect on each input parameter. In each case, a (+) indicates that use
of the practice will increase the level of that parameter in the predictive
equations, a (-) indicates a decrease in that parameter, and a blank
indicates that the practice will not affect the level of that parameter.
151
-------
TABLE A-9. INPUT DATA REQUIREMENTS, COMPUTATIONAL PROCEDURE, AND RESULTING
LOADING ESTIMATES BY USE OF THE PREDICTIVE EQUATIONS
POLLUTANT:
INPUT DATA
PARAMETERS:
COMPUTE
EDGE-OF-FIELD
LOADINGS:
ESTIMATE
NUTRIENT DELIVERY:
SOLID-PHASE N
Total soil N concentration
SSN (mg/kg)
USLE factors:
K
Slope (%)
Slope length (m)
C
P
Enrichment ratio for soil
N in eroded soil
1) Compute average annual
soil erosion by eq. 10
(T/ha-yr)
2) Compute average annual
total N loss in eroded
soil by eq. 11 (kg/ha-yr)
Compute sediment delivery
by eq. 16 or 18
SOLID-PHASE P
Total soil P concentration
Ssp (mg/kg)
USLE factors:
K
Slope (%)
Slope length (m)
C
P
Enrichment ratio for soil
P in eroded soil
1) Compute average annual
soil erosion by eq. 10
(T/ha-yr)
2) Compute average annual
total P loss in eroded
soil by eq.
12 (kg/ha-yr)
Compute sediment delivery
by eq. 16 or 18
POLLUTANT:
INPUT DATA
PARAMETERS:
DISSOLVED N
Inorganic N in fertilizer
FN (102 kg/ha-yr)
Soil organic N Sw (10
kg/ha-10 cm) N
DISSOLVED P
Available P in fertilizer
Fp (kg/ha-yr)
Soil available P Sp (kg/ha-
10 cm) v
Curve number CN
II
Curve number CN
II
Soil available water
capacity AW (cm/cm)
Soil hydrologic group
Average annual
precipitation Pr (cm)
Soil available water
capacity AW (cm/cm)
Soil hydrologic group
Average annual
precipitation Pr (cm)
o
Soil bulk density p(g/cm )
Soil pH
Soil clay content %C (%}
(cont.)
152
-------
TABLE A-9. (Continued)
COMPUTE
EDGE-OF-FIELD
LOADINGS:
ESTIMATE
NUTRIENT
DELIVERY:
1) Compute annual runoff by
eq. 1 & 3 (cm/yr)
2) Compute annual perco-
lation by eq. 2 & 4
(cm/yr)
3) Compute average dis-
solved N concentration
in percolation by eq.
6 (mg/1)
4) Compute average dissolved
N concentration in perc-
olation by eq. 6 (mg/1)
5) Compute average annual
dissolved N loadings in
runoff and percolation
by eq. 13 & 14 (kg/ha-yr)
Average annual stream
dissolved N loading
= runoff N plus
percolation N
1) Compute annual runoff by
eq. 1 and 3 (cm/yr)
2) Compute 3 from pH and %C
by eq. 8 (I/kg)
3) Compute ypfrom 3 and p
by eq. 9 (dnless)
4) Compute average dissolved
P concentration in runoff
by eq. 7 (mg/1)
5) Compute average annual
available P loading in
runoff by eq. 15 (kg/ha-yr)
Average annual stream
dissolved P loading
- runoff P
153
-------
oo
t^>
o
D-
oo
a:
QC
0 0-0
I— i. -r- -r- -r-
CD T- O) 4-> -l-> to
E CO 4J q- ro ro CU
•r- E O S- S-
S- T- T3 -O v- O
3 O CU S- 3 3 Q. O 5-
E i- XJ T3 ro O O
O CU CU CU E
o i— cc a£ i—i
1 1
1 1
~*~ '
-"" **
' '
to
1 to to E
E i- 0.0
O>
O O -r- U ro
1— 4->
E S- ••- i- O
O O +-» CU i-
•r- 4- S- >
4-> CU O "O
3 O) q- O O)
4-> i- tO
•r- 3 i— S- ro
4-> E ro CU -Q
10 rO CJ 4-> I
-Q E "- E -0
3 -r- O
00 S 00
sx
E
(O
S -D
CU
(13 ^—
•o -^
E +->
ro
S
•> O
CU S-
to i
ro -4->
cu ^:
i- cn
u -i-
cu ro
-a s-
i ^
ro to
to ro
0)
4-> O
ro 4->
O
•r- -a
-a cu
E i.
•r- rO
D.
I E
O
ro O
r. (/)
CU (O
O
•i- CU
1 ^ f
O ro
ro
S- 10
o CU
O
a> •!-
c~* ^_J
4_) (j i_
ro CU
^: i- >
+-> a. o
•r- O
^ s-
S- et CU
CU -I-J
4-* E
CU • •!-
E S- 2
rO CU
S- 4-> O
(O CU E
Q- E
ro -E
CU i- 4J
.c ro -i-
4-> CX S
E CU CX
•r- .E O
4-> S-
CU O
to E
rO O T3
CU CU
i- 4-» N
O CJ •!-
E CU i—
•r- 4- -r-
M- 4-)
E CU i-
rO CU
0 4-
tO E
cu >,
4-> to •—
ro (U i—
U 4-> ro
•r- ro O
T3 O -i-
E -I- E
• i— ~O CU
E -E
+ •- 0
•
F—
to
a>
u
•>-
o
ra
i-
D-
E
cu
E
O)
cn
ro
E
CO
E
a.
o
S-
CJ
O)
to
0)
£
>^
<^i
•a
cu
-f >
o
cu
q-
q«
ro
E
~"^
CU
i-
rO
to
E
O
•r-
-)._>
ro
73
cr
cu
cu
>
• r-
^_)
O
•r—
T3
a>
i.
Q-
E
• r-
to
i-
a>
4^
cu
E
ro
i-
ro
Q-
^
O)
-E
I '
O
^~
^_
^^
•
OJ
154
-------
to
c
5
-Q
Q
CO
S-
o
E
S-
0
0)
2 O
0 E
/1 1
r~~ UJ
Q_ C£
CD (/)
c o>
•i — Z3
e T~1
>- \J
f^ •!—
CO
OJ
o:
5"£!
O O
Ofl 1
\U
r— CC
Q-
l/l
r— CU
frt "^~j
ly -~.t
U. t-
<1>
o:
+j
OJ
2;
o to
r— CD
D. 3
T:
^^ t/)
03 O)
u. o:
^~
p™"
h^
i
0
c~
O
o>
a;
Louni— cvjooocsjr— ooo^j-coc3-i<£>CTi<^-LncMr~.Lnioroio<4D^-i—
Lnirja^LnLnLnLnu^^Lnuni^LOLnLOL^in^LnLOLOLnLnLnunLn
i i i i i i i i i i i i i i i i i < i i i i i i i i
CSJ C\J ^^ O1 O1 LO CO ^^ CVJ LO i^ ^" VO CNJ ^^ f") C\J ^~* ^j" C\J CSJ O^ OO CO CZ5 ^O
<^cncxic\)c\jCMC>jcjc\jc^coco<^^oooo(^^cv7(v7focvjO O^ CO CXJ LO C\J LO CO CO CO n~" CO C\J ^^3 ^^" ^^" r~ ^if* ^^~ i~™ • CO
LO LO *^^~ LO LO *«J* *^~ "5^ ^t" ^^" LO LO LO LO LO LO LO ^.O LO LO LO LO LO LO LO "^1"
i i i i i i i i i i i i i i i i i i i i i i i i i i
i-o in co r~- r~™ r^s co co r— - LO oo r^ (~"i f~~i co CM LO LO r**» LO *^ c\j to LO co co
•=J-^-oo«a-^j-co^i-cococo<-^LOLOLo^t^-Lo^-i — coi — uococoi — <*«d-i — oo
•*«d-co«d-^f-co^i-cococO'=i-^-^}-^-^i-^-^-Lo«*^-^-^«5i-co
i i i i i i i i i i i i i i i i i i i i i i i i i i
LOLOOCMl— CTlr— CTlLOI^i— «*^J-^^C\JCOI^CMOC\l
JO
.
to
cu
3
•r-
to
cu
%-
CD-
CD
<*.
01
E
•p-
a.
to
•4-
o
to
CL>
>
CU
"~
J-
O5
!E r^
Csl
E C
03 O
•r~
2 CT
o cu
o o
M-
>,
4-> CD
> JD
•r- 03
0 ^
!3 ITS
O 03
S- E
a. 3
JE 03
f-rv j\
^^7 T-^
•r- 03
-E TD
"O i—
03 O3
•S E
0 •!-
"~ S-
ai o
to
d.*»—
ai •»->
S- 3
-Q
to -i—
O)4->
E 10
03 -r-
i:ci
•^-^,
155
-------
TABLE A-ll. (cont.)
No tillage
Strip-till,
straight row
Strip-till
on contour
Turn-plow
Corn with
Winter Cover
0.07 - 0.15
0.12 - 0.21
0.11 - 0.18
0.28 - 0.43
Continuous
Hay
Grass and 0.004 - .010
legume mix
Clover
Alfalfa
0.015
0.020
'Ranges represent low and high productivity or low and high levels of spring
crop residues
156
-------
PREDICTION OF SOLID-PHASE EDGE-OF-FIELD NUTRIENT LOADINGS
Solid-phase nitrogen and phosphorus losses in runoff are dependent on
the detachment and transport of soil particles from the field. The primary
mechanism for soil detachment on the field is rainfall impact, although small
amounts of erosion may result from snowmelt movement.
Gully and channel erosion may also contribute to stream sediment
loading; however, these are off-field effects, and form part of the sediment
delivery process. The present discussion concerns only rainfall erosion.
Average annual soil erosion to edge-of-field under a specific practice
is predicted through application of the USLE (Wischmeier and Smith, 1978):
A = 2.24 R K LS C P (10)
where A = average annual soil erosion (T/ha-yr)
R = rainfall-runoff factor
K = soil erodability factor
LS = slope-length factor
C = crop cover and management factor
P = support practice factor
Methods for obtaining field-specific values of R, K, LS, C, and P are given
in Wischmeier and Smith (1978). The crop cover factor C may also be obtained
from the generalized values in Table A-ll. These factors were determined by
estimating an average time period for each crop stage (fallow, seedbed, crop
establishment, crop development, maturing, and post-harvest) for each crop
and region, based on the average dates of plant and harvest (U.S. Department
of Agriculture, 1972) and the crop development rates given in Table A-5. For
each crop stage, a C factor was picked for both low and high productivity (or
low and high levels of crop cover). These C factors were multiplied by the
average fraction of yearly erosion which occurs in each crop stage, and the
products were summed over all crop stages to obtain an average annual C
factor which is reflective of both crop management and the distribution of
erosive rainfall during the year.
In addition to soil erosion estimates, prediction of solid-phase
nutrient loss requires an estimate of the nitrogen and phosphorus concentra-
tions in the eroded soil. The level of these nutrients will depend on the
parent material, additions to fertility through manure applications or
plowing under crop residues, or other factors. It is therefore impossible to
estimate these nutrient levels for specific field without some form of
sampling, either directly in the field, or by monitoring nutrient concentra-
tions in sediment carried in runoff water. If soil sampling is used, the
soil nutrient concentrations should be multiplied by an appropriate enrich-
ment ratio because eroded soil particles are commonly lighter and higher in
organic matter and clay than the remaining surface soil, and will therefore
contain higher concentrations of both solid-phase N and P. Suggested ranges
of the enrichment ratio are 2.0 and 4.3 for soil N and 1.5 to 3.4 for soil P
(McElroy _et _al_., 1976). Specific values of 2.5 for soil N and 2.0 for soil P
are used in the examples in Section 4.
157
-------
Edge-of -field loadings of N and P in eroded soil are computed by
combining the USLE with soil nutrient concentrations and enrichment ratios
for N and P:
SN = 0.001 ERN A SSN
(11)
SP = 9.001 ERp A SSP
(12)
where SN
SP
average annual solid-phase N loss in eroded soil (kg/ha-yr)
average annual solid-phase P loss in eroded soil (kg/ha-yr)
ERp = enrichment ratio for soil N and P (dmless)
average annual soil erosion (T/ha-yr)
concentration of organic N in surface soil (mg/kg)
concentration of solid-phase P in surface soil (mg/kg)
PREDICTION OF DISSOLVED EDGE-OF-FIELD NUTRIENT LOADINGS
Through repeated application of the CNS model (see Section 2), the
following predictive equations for dissolved nutrient losses in runoff and
percolation were derived for the crops and practices listed in Table A-8:
Runoff and Percolation Volume
%R = a CN + b
%P = a CN + b
(1)
(2)
where %R = average percent of annual precipitation appearing as direct runoff
%P = average percent of annual precipitation which percolates below 30
cm.
CN = - where CNir = crop curve number for average soil
100 - CNH u
moisture (antecedent moisture condition II)
a, b, a , b = constants for each crop, soil, and cropping practice
and,
(3)
P -
IDD
where R
P
Pr
average annual direct runoff (cm)
average annual percolation (cm)
average annual Precipitation (cm)
158
-------
Dissolved Nutrient Concentrations in Runoff and Percolation
KRN = aO + alCN + a2FN + a3sN + a4FNCN + 35SNCN (5)
KPN = bg + t>iCN + b£FN + b3$N + b4FNCN + bsS^jCN (6)
A A A
= yp (Co + ciCN + C2Fp + C3$p + C4FpCN + C5$pCN) (7)
where Kp^ = average concentration of dissolved N in runoff (mg/£)
= average concentration of dissolved N in percolation (mg/Jl)
= average concentration of dissolved P in runoff (yg/A)
F|\j = annual fertilizer and manure inorganic N input (102 kg/ha-yr)
SN = soil organic N in the surface 10 cm (103 kg/ha)
Yp = adjustment factor for soil P adsorption capacity (dmless)
Fp = annual fertilizer and manure available P input (kg/ha-yr)
Sp = soil available P in the surface 10 cm (kg/ha)
a-,-, b-j, c-j = constants for each crop, soil, and cropping practice
84.5
where p = soil bulk density in the surface 10 cm (g/cm3)
3 = soil available P adsorption coefficient ffl/ 9)
mcj / X
2
and 3 = 5.1 + 2.2 (%C) + 26.4 (pH - 6.0) (8)
where %C = percent clay content (soil particles < 0.002 mm in diameter) in
the surface 10 cm
pH = average pH of surface 10 cm
Dissolved Nutrient Loading in Runoff and Percolation
Average dissolved nutrient loading to edge-of -field is found by com-
bining the runoff, percolation, and nutrient concentration estimates:
DNR = 0.1 (R) (KRN) (13)
DNP = 0.1 (P) (KPN) (14)
DPR = 0.0001 (R) (KRp) (15)
159
-------
where DNR = average annual dissolved N loading in runoff (kg/ha-yr)
DNP = average annual dissolved N loading in percolation (kg/ha-yr)
DPR = average annual dissolved P loading in runoff (kg/ha-yr)
IMPLEMENTATION OF THE NUTRIENT LOADING PREDICTIVE EQUATIONS
The procedure to be followed to implement the predictive equations may
be broken down into four steps for dissolved nutrient loadings and three
steps for solid-phase nutrient loadings.
Dissolved Nutrient Loadings
1) Obtain values for input parameters (CM, Pr, FJ\J, Fp, SN, Sp, AW, p, %C,
pH)
a) Curve number factor CN is equal to 100/(100-CNjj), where CNjj is the
SCS curve number for the crop for average moisture conditions (AMC
II). Curve number is obtainable from Volume II of Stewart et a1.,
(1976), or Ogrosky and Mockus (1964).
b) Fertilizer application levels F^ and Fp include all inorganic N and
crop available P applied per year. If manure is applied, the amount
of inorganic N in the manure that is volatilized before incorpora-
tion in the soil must be subtracted from F^. Average fertilization
levels, which may be used in the absence of specific field data, are
given in Table A-12 (Beaton and Tisdale, 1969).
c) Soil organic N and available P levels must be determined by soil
sampling in the field under study.
d) All other input parameters (Pr, AW, p, %C, and pH) are constants for
the particular field under study. Values of p, %C, and pH are
needed only if an estimate of average dissolved P loading in runoff
is desired. Soil bulk density, soil available water capacity, and
clay content are usually available from soil surveys, whereas soil
pH should be determined by sampling in the field.
2) Compute average annual runoff and percolation
Given field-specific values for CN and Pr, equation 1 through 4 may be
used to estimate average annual runoff and percolation given the appropriate
values for a, b, a , and b from Appendix III. The coefficients in Appendix
C are chosen by crop, time of plowing, soil hydrologic group, geographic
region, and soil available water capacity (AW, as a percent of soil volume).
The appropriate table may be most easily located by reference to the list of
tables at the beginning of Appendix III. Average runoff and percolation
estimates for available water capacities intermediate to those listed in
Appendix III should be determined by interpolation.
3) Compute average dissolved N concentrations in runoff and percolation and
dissolved P concentrations in runoff
Nutrient concentrations are determined in a manner similar to that for
runoff and percolation volume. Given values for CN, F^, and S^, dissolved
160
-------
00
Q.
O
(_)
CC.
O
•-D
e^£
o;
o
:r
CD
^5
1 1 1
i i
i
t — i
LU
U_
l— i
oo cr>
•Z CO
O O1
1 — 1 1 —
<:
c_> cu
1 | ,
_1 n3
Q- TD
Q- (/)
T3
rv* c~
O rO
Q. E
oo o
O -I-3
1 fQ
IX -
Q
1 ^3
=C I—
^D OO
z in
• LU
LU
:> z.
CM
1
LU
1
GO
•4-^
(C
^: CD
•i- 3
^
^~
E CD
i- +->
O E
is
l/l
OJ
JD
oo
D- JC
O
o:
c_>
O)
E 0)
i- to
O i —
O •!-
oo
e
E T-
s- ro
0 1-
C_) CD
o
i — i
CD
LU
CC
LOLO. .LOLOOOOOOO . OLDLDO
D- CMCM OJCXJLO'd"^^'*'^ 'tOLO^LT}
oo OOOUILOLOOO LOUOOLO
~f^ c~~t f~*t i i i co co c? LO LO co (^ ^- i i r*^- cTi oo ^«
LO LO LO f~*l ("^ f^ f"s f"~) f~^ c i c~^ LO LO LO LO f~^ ^__j LO LO
CD CO CD LO CD LO LO CD CD CD LO LO LO LO CD CO LO CO CO
OOLOOOOOLO. . ,LO. . .OOLOO
Q. CMCMCMCMCOCMCMCM ' ' *CM ' ' 'OOOO<=J-CO
LOLOOOLOOOO, . tO. . .OOOO
^^. CO CO l(vj" cj* ^J" ^^ i**~ CO 111 u~j iii ^^. u~j JY"J fyj
l-O LO LO C""l C~^ LO LO LO LO LO LO LO LO LO LO LO UO LO LO
D. i — i — CMCMCOi — i — CMCMCMC\JC\JCMC\JC\l*3-OOrOCO
OOLOOOOOOOOOOOOOLOLOOO
^ CO CO O CO CO O O •vt" i — r— OJ i — CM *^~ ^ r^ LO LO CO
LOLOOOLOLOLOLOOOLOOOLOOLOLOOO
O- i — i — CMCMCMi — i — CXICMCMCVJCMCMOJCMCMCXJCOCO
LOLOLOLOLOOOOOOLOOOOOOOOO
r— CMCO^I-LOCOI^COCTlOr— CMOO 1 1 CO CO 1
CM r-~ CM CM
o o , . ,000
CM CM CO CO CO
o o , . .000
III ^_!
. . LO O O LO O .
1 ' CM CM •* CM «* '
. . LO LO O O , .
1 ' CM CO LO CO ' '
COCMCOCMLOCMCMLO
OOOLOOOLOO
oooooooo
COCMCOUDCMCMCO«^-
OOOLOOOOLO
Oi — CMco^j-iocor^
CM CM CM CM C\J CM CM CM
161
-------
N concentrations in runoff and percolation may be computed from equations 5
and 6 given the appropriate values of a and b from Appendix III. Dissolved
i i
P concentrations in runoff may be computed from CN, Fp, Sp, p, %C, and pH
using equations 7, 8, and 9 and the appropriate values for c-j from Appendix
III. The coefficients in Appendix III are selected by crop, time of plowing,
time of fertilization, soil hydrologic group, geographical region, and soil
available water capacity (AW).
4) Compute average annual dissolved N and P loadings
Average annual dissolved N loadings in runoff and percolation and dis-
solved P loadings in runoff are found by combining average annual runoff and
percolation estimates from step 2 with the average nutrient concentrations
from step 3, according to equations 13, 14, and 15.
Solid-phase Nutrient Loading in Runoff
1) Obtain values for input parameters (S$N, S$p, R, K, LS, C, P, ER|\(, ERp)
Soil concentrations of solid-phase N and P (5$^, S$p) should be obtained
by soil sampling in the field or by analysis of eroded soil. Factors for the
USLE (equation 10) may be taken from Wischmeier and Smith (1978), or
generalized C factors may be taken from Table 11. Enrichment ratios for soil
N and P are variable, but may be chosen as ER^ = 2.5 and ERp = 2.0.
2) Compute average annual soil erosion
Average annual soil loss to edge-of-field is computed using the USLE and
the values derived for R, K, LS, C, and P for each study field.
3) Compute average annual solid-phase nutrient loss
Average annual solid-phase N and P loss to edge-of-field is found by
substituting the computed value for average annual soil erosion A from step 2
and the soil nutrient concentrations and enrichment ratios from step I into
equations 11 and 12.
Examples of each of these procedures are given in Section 4 of this report.
DELIVERY OF CROP NUTRIENTS FROM EDGE-OF-FIELD TO SURFACE WATERS
Once estimates of edge-of-field nutrient losses have been obtained, it
may be necessary to determine the nutrient fraction which will reach surface
waters through direct runoff or baseflow. The various pollutant transport
mechanisms may be separated into 1) dissolved pollutant transport in direct
runoff; 2) sediment and solid-phase pollutant transport in direct runoff; and
3) dissolved pollutant transport in groundwater (Figure A-l). Practices
designed to control pollutant transport may affect one or more of these
mechanisms.
162
-------
Solid-Phase Pollutant Transport in Direct Runoff
After leaving the field in surface runoff, sediment and associated
pollutants may be removed from runoff by filtration and deposition. Sediment
may be deposited in grassed waterways, field borders, or buffer strips, or in
forest, pasture, and cropland downslope of the field. The effectiveness
(percent reduction in sediment loading) of sediment control practices will
depend on the specific application of that practice, and is not easily
predicted from secondary information. Sediment loading reductions through
the use of individual off-field practices must therefore be determined on a
local level, by observing the effectiveness of each candidate practice in an
existing management scheme or through monitoring of sample fields.
Sediment deposition on intervening land surfaces may be approximated by
a simple distance-to-stream function if the location of the field relative to
the nearest permanent water body is known, or by a watershed drainage density
model in the absence of field location data. The first method requires the
computation of a delivery ratio (Sd-j) for each field in a study area, and
combines with the USLE to estimate the sediment contribution of each field to
the receiving waters:
YT = Sdi Ai (16)
where Y-j = in-stream sediment loading from field i (T/ha-yr)
A-,- = sediment loading to edge-of-field for field i (T/ha-yr)
Sd-j = sediment delivery ratio for field i (dmless)
-0.36
and Sdi = 2.5di (17)
where d-j = distance from center of field i to receiving water (m) (Smith, et_
al., 1979). The second method computes an average watershed delivery ratio
as a function of drainage density (the ratio of the sum of permanent channel
lengths in the watershed to total watershed area), and assumes a relatively
homogenous soil texture in the watershed:
Yi = Sd Ai (18)
where Sd = average watershed sediment delivery ratio (dmless)
Average sediment delivery ratio "ScT is obtainable from Figure A-7 (McElroy et
n n
al., 1976), given the reciprocal of drainage density in km /km or mi /mi and
the predominant soil texture in the watershed.
Dissolved Nutrient Delivery in Direct Runoff
The fraction of runoff which reaches surface waters depends primarily on
the location of the field relative to the nearest conveyance channel. If the
field drains directly into a stream or unimproved drainage ditch, we may
assume that all runoff will enter the surface water system. If the field is
separated by forest, pasture, or other cropland, some or all of the runoff
163
-------
o
o
III I I I I I
o
o
o
oo
o
oo
LU
Q
UJ
en
z
UJ
o
o
o
oo ^->
D
r^.
>- cr>
Qi r—
2: o
I—I QL
Q _l
UJ UJ
oo
oo o;
H- x
-------
may be captured in the intervening area through ponding and subsequent
infiltration. However, diffuse runoff from the field may quickly coalesce
into small intermittent streams of higher velocity, which may reach a
permanent channel without significant reduction in flow volume. A reasonable
assumption is that 100% of direct runoff will reach a surface water body,
unless diversions or detention ponds are used to redirect the flow to a place
where infiltration can occur. Dissolved nitrogen is generally conserved in
runoff, and hence all dissolved N lost from the field in runoff will enter
the surface water systems. Dissolved phosphorus may be partially adsorbed by
suspended sediment during transport or after deposition in the stream. The
dissolved P trapped by sediment may reappear in the stream if the deposited
or trapped sediment is detached by a later storm event. Complete delivery to
the stream of dissolved P in runoff is a conservative assumption, although
this is not a highly reliable estimate.
Dissolved Pollutant Transport in Groundwater
Percolation below the root zone will travel to the groundwater table,
and may appear as baseflow in streams or lakes. The transport process may
occur over a long period, during which time the nutrients carried in
percolation may be diluted or changed in form. Dissolved P in percolation
will reach equilibrium with the adsorbed phase in deep soil layers before
reaching surface waters. The equilibrium concentration in groundwater
reaching the stream or lake will therefore be close to background levels.
Conversely, dissolved N is essentially conserved in groundwater, and field
losses of dissolved N may reappear in baseflow. Although the eventual
destination or timing of reappearance of percolated N from each field may not
be known, the total watershed N flux reaching surface waters in baseflow
should be well predicted by the sum of the individual field loadings in
percolation, as determined by the predictive equations for dissolved N in
percolation.
165
-------
SECTION 4
DEMONSTRATION OF THE USE OF THE NUTRIENT
LOADING PREDICTIVE EQUATIONS
Implementation of the nutrient loading equations is best demonstrated
through examples of their use. In this section, two situations are consid-
ered: first, runoff and nutrient loading estimates are computed for a
hypothetical cropping scheme in three different geographic regions; and
second, the predictive equations are utilized to predict average annual
nutrient loadings for the Honey Creek watershed in northern Ohio.
LOADING ESTIMATES FOR A HYPOTHETICAL CROP SITUATION
Consider a straight-row corn crop on a soil of hydrologic group C,
fertilized in the spring and turn or moldboard plowed after harvest in the
fall. The crop practice and soils data necessary to use the predictive
equations are given in Table A-13. In order to demonstrate the variation in
nutrient loading between locations, loading estimates are computed for this
crop in three geographic regions: southern Minnesota (region 4), central
Texas (region 10), and Pennsylvania (region 22). Within each region, two
management practices, contouring and no-tillage, are analyzed for their
effects on nutrient loadings as compared to base conditions.
Base Conditions (straight-row, fall-plowed, spring-fertilized)
Solid-phase Nutrient Loadings--
Solid-phase nutrient loss is predicted by combining the USLE (equation
10) with soil nutrient concentration data, according to equations 11 and 12.
Rainfall-runoff factor R as obtained from Wischmeier and Smith (1978) is
approximately equal to 125 for region 4, 250 for region 10, and 125 for
region 22. Crop cover factors (C) for each region are obtained from Table
A-ll, for fall plowing with residues removed, assuming average crop cover.
Average annual soil loss is then computed as:
TABLE A-14. COMPUTATION OF AVERAGE ANNUAL SOIL LOSS UNDER BASE CONDITIONS
Region
2.24
LS
A
4
10
22
(2.24) (125)(.25) (,354)(.45)(1 .0)
(2.24) (250)(.25) (.354)(.40)(1 .0)
(2.24) (125)(.25) (.354)(.46)(1 .0)
= 11.2 T/ha-yr
= 19.8 T/ha-yr
= 11.4 T/ha-yr
166
-------
TABLE A-13. FIELD CHARACTERISTICS FOR SAMPLE CROP
FIELD CHARACTERISTIC
VALUE
Soil total phosphorus content
Soil organic N level
Soil available P level
Fertilizer N application
Fertilizer P application
a/
PARAMETER VALUE
Soil erosivity
Slope length
Percent slope
Soil available water capacity
Percent clay content
Soil pH
Soil bulk density
Soil total nitrogen content
0.25
61 m
3%
.15 cm/cm
10%
6.5
1.5 g/cm3
1330 mg/kg
K =
LS
AW
%C
PH
p =
3 =
YP
SSN
0.25
= 0.354
= 15
= 10
= 6.5
1.5
33.7
= 1.67
= 1330
333 mg/kg Ssp = 333
2000 kg/ha-10 cm SN = 2.0
50 kg/ha-10 cm Sp = 50
100 kg/ha-yr FN = 1.0
25 kg/ha-yr Fw = 25
a/ In most cases solid-phase N and P must be approximated by total soil N
and P.
167
-------
Average annual solid-phase N and P loadings are then computed by equations 11
and 12:
TABLE A~15 COMPUTATION OF AVERAGE SOLID-PHASE NUTRIENT LOADINGS
Region Nitrogen:
4
10
22
Region Phosphorus:
4
10
22
.001 ERN A, SSN
(.001) (2.5) (11.2) (1330)
(.001) (2.5) (19.8) (1330)
(.001) (2.5) (11.4) (1330)
.001 ERp Ab Ssp =
(.001) (2.0) (11.2) (333) =
(.001) (2.0) (19.8) (333) =
(.001) (2.0) (11.4) (333) =
SN
= 37.2 kg/ha-yr
= 65.8 kg/ha-yr
= 37.9 kg/ha-yr
SP
7.5 kg/ha-yr
13.2 kg/ha-yr
7.6 kg/ha-yr
Dissolved Nutrient Loadings—
From Stewart et_ _a1_. (1976) crop curve number equals 88 (CN = 8.33) for a
straight-row crop in poor hydrologic condition (residues removed) on a C
soil. Average annual precipitation for the three locations was set equal to
the levels used in the CNS model simulations. Because the crop is fall
plowed, Table (A)III-IO contains the appropriate values for a, b, a1, and b1
Nutrient coefficients are obtained from Table (A)III-ll, for a fall- plowed
corn crop fertilized in the spring. The following table shows the steps to
be followed to compute average annual runoff and percolation from the data
given and equations 1 through 4:
TABLE A-16. COMPUTATIONAL PROCEDURE FOR DIRECT RUNOFF AND PERCOLATION
Runoff
region
4
10
22
Pr
(cm/yr)
64.1
59.2
92.0
ti
8.33
8.33
8.33
a
1.60
1.33
1.32
b %R
12.67 26.
10.63 21.
21.42 32.
R
(cm/yr)
0 16.7
7 12.8
4 29.8
-1
-1
-1
a1
.44
.01
.32
Percolation
b1
30.23
20.73
39.79
%P
18.2
12.3
28.8
P
(cm/yr)
11.
7.
26.
7
3
5
168
-------
Average annual dissolved nutrient loadings for base conditions are computed
similarly, using the crop practice and soils data from Table A-12, the coef-
ficents a-j, bj, and c-j from Table (A)III-ll, and equations 5-7 and 13-15:
TABLE A-17. COMPUTATIONAL PROCEDURE FOR DISSOLVED NUTRIENT LOADINGS
Runoff N
Percolation N Runoff P
R P KRN RN KpN PN KRp RP
region (cm/yr) (cm/yr) (mg/1 ) (kg/ha-yr) (mg/1 ) (kg/ha-yr) (yg/1 ) (kg/ha-yr)
4
10
22
16.7
12.8
29.8
11.7 4.8 8.0
7.3 6.5 8.3
26.5 2.2 6.6
55.7 65.2 116.3 0.19
99.0 71.7 106.3 0.14
20.9 55.4 107.5 0.32
Contouri
ng as
a Management Practice
Contouring will affect both soil loss levels and dissolved nutrient
losses. The only effect of contouring on soil loss is through the change in
the support practice factor P, which is equal to 0.5 for contouring on a 3%
slope. Soil loss and resulting solid-phase nutrient loadings are therefore
cut in half for all regions:
TABLE A-18. SOIL AND SOLID-PHASE NUTRIENT LOSS WITH CONTOURING
region
4
10
22
Soil Loss
(T/ha-yr)
5.6
9.9
5.7
Solid-phase N Loss
(kg/ha-yr)
18.6
32.9
19.0
Solid-phase P
(kg/ha-yr)
3.8
6.6
3.8
Loss
The effect of contouring on dissolved nutrient loss is expressed through a
change in the curve number from 88 to 84 for contouring (CN = 6.25). The
same values for a, b, a', b', a-j, b-j, and c-j are used to predict runoff,
percolation, and nutrient loadings:
TABLE A-19. COMPUTATION OF DISSOLVED NUTRIENT LOADINGS FOR CONTOURING
region
4
10
22
R
(cm/yr)
14.5
11.2
27.3
P
(cra/yr)
13.6
8.5
29.0
Runoff
N
Percolation N
Runoff P
KRN RN KpN PN KRp RP
(mg/1) (kg/ha-yr) (mg/1) (kg/ha-yr) (yg/1) (kg/ha-yr)
6.2
6.1
1.9
9.0
6.8
5.2
50.7
89.0
19.8
69.0
75.7
57.4
113.7
104.7
106.1
0.16
0.12
0.29
169
-------
No-Tillage as a Management Practice
Tillage elimination will affect both soil loss and dissolved nutrient
loadings. In this case, soil loss is decreased through a reduction in the
crop C factor. Therefore, both soil and solid-phase nutrient losses will be
reduced by the ratio of the no-tillage C factor to the fall plowing C factor.
TABLE A-20. SOIL AND SOLID-PHASE NUTRIENT LOSSES UNDER REDUCED TILLAGE
region
4
10
22
Soil loss
(T/ha-yr)
2,7
5.9
3.0
Solid-phase N loss
(kg/ha-yr)
9.1
19.7
9.9
Solid-phase P loss
(kg/ha-yr)
1.8
4.0
2.0
Dissolved nutrient loss is computed using the predictive equation
coefficients in Table (A)III-l and (A)III-2 for^a no-tillage corn crop
fertilized in the spring. Curve number factor CN is equal to 8.33. Estima-
ted runoff, percolation, and dissolved nutrient losses are presented below:
TABLE A-21. RUNOFF, PRECOLATION, AND DISSOLVED NUTRIENT LOADINGS FOR REDUCED
TILLAGE
R
Runoff N
l\r\kt
RN
Percolation N
PN
Runoff P
RP
PN
-------
Q_
O
OL
O
LLJ
1
Q_
^-
i — i
a
LU
a;
a.
oo
oo
o
i
f—
2:
UJ
KH
a:
H-
^
-*
~*^
a
2:
a;
1 1
o
>-
cn.
00
.
CM
CM
=£
L-U
— 1
CO
=C
1—
CO
O)
CO
CO
o
_J
0)
CO
rO
~c
Q.
X)
•r~
o
oo
CO
cu
co
CO
o
, 1
-o
a>
j_^
0
CO
CO
•1—
Q
1?
>.
•-O
0 J*
•"*" fO
_c
cn
_M
.^ x
i.
^^^
>1
ro
^ jz
cn
^^
V '
CX S-
«*- ^
It- (0
o .c:
c: ~-^
zs cn
rv ^;
2:
TD i-
OJ >)
•M 1
(D fC
i — _C
O ~\
O cn
i- -^
0) — '
Q_
i_
Z >,
1
M- ra
<<- -C
O ~~-
c cn
Z3 -V
a: —
CO • — -
CO S-
O >i
" i
•>
r— "^
0 E
o o
CD
a_
C|_ , ,
M- S-
i 1*
>~- £^
3 (J
o; ^_-
fd
01
u
-J3
o
tocnoo
i—i—i— CM CM OO
l~-~Lnr— ODCMCMCOOOOO
to^J-oocMi — ocnr-^cM
Q; C£ C£ C£. C£. C£.
OOC_)OOOOC_)OOOOC_30O
Q.Q.I — D_ D_ I — Q-O-I —
LJ_U_CC:U_U_CC:LI_IJ_Q:
'd- O CM
t— CM
-o
a>
^
0
-(->
c
o
o
II
CJ
r>
S
o
4-J
^:
cn
rO
S-
-I-J
CO
II
a:
oo
O)
en
(O
1
•r™
+-)
•o
CU
u
13
XI
OJ
i-
II
£
A
2
0
a.
^^
^_
rO
ti-
ll
a.
u_
^^^
fO
171
-------
CD
UJ
UJ
CC
O
UJ
z
O
X
TD
O)
ro
CD
OJ
O
O)
O
00
O)
S-
172
-------
are computed for each subwatershed, and are then combined to determine total
watershed losses.
Soil N levels are derived from soil organic matter data obtained from local
soil surveys (Soil Conservation Service, 1975), assuming a 20:1 carbon to
nitrogen ratio (therefore organic matter is 5% nitrogen). Available P and
solid-phase P data is averaged over all soil groups, since specific data was
only available for three or four soil groups. Fertilizer N and P levels are
set equal to the recommended levels for each crop in Ohio (Beaton and
Tisdale, 1969), as given in Table A-12. Consequently, the nutrient loss
estimates computed in this example are rough approximations, although the
reduction in nutrient loading by the use of alternative practices should be
well predicted.
Base conditions for the watershed are defined as fall plowing for corn
and soybeans (except when grown with a winter cover crop, in which case
no-tillage is assumed), and all crops planted straight-row. Hay is grown
continuously without tillage. The areas of the various crops and rotations
in the four subwatersheds are given in Table A-24. Nutrient loadings were
estimated by applying the predictive equations to each crop in each rotation,
and averaging over all crops in the rotation to obtain average annual
loadings.
In addition to base conditions, three management options were analyzed
for their effects on nutrient loadings to edge-of-field: 1) contouring on all
crops 2) spring plowing rather than fall plowing for corn and soybeans, and
3) no-tillage rather than conventional tillage for corn and soybeans.
Nutrient loadings were predicted for both spring and fall fertilization, to
represent the expected range of loadings in the watershed. Contouring
affects the value of CN for each crop, while variations in tillage requires a
shift from one set of predictive equation coefficients to another, without
any change in the input parameter values. The additional data necessary to
implementing the predictive equations is as follows:
average precipitation = 95.3 cm/year
fertilizer N applications: corn = 150 kg/ha-yr (F^ = 1.5)
soybeans = 5 kg/ha-yr (F^ = 0.05)
hay = 50 kg/ha-yr (FN = 0.5)
corn with winter cover = 200 kg/ha-yr
(FN = 2.0)
fertilizer P applications: corn = 30 kg/ha-yr (Fp = 30)
soybeans = 20 kg/ha-yr (Fp = 20)
hay = 30 kg/ha-yr (Fp = 30)
corn with winter cover = 60 kg/ha-yr
(Fp = 60)
Solid-Phase Nutrient Loadings
Average annual soil loss by subwatershed and soil group as determined by
the USLE is given in Table A-25. These soil loss data are combined with the
173
-------
soil nutrient data in Table A-23 to provide an estimate of average annual
solid-phase nutrient loadings to edge-of-field in each subwatershed under
base conditions. Enrichment ratios for soil nutrients were set at 2.5 for
soil N and 2.0 for soil P. A similar procedure was followed for nutrient
loading under each alternative management option, given the expected average
annual soil loss under each practice. Table A-25 contains the estimated
nutrient fluxes for each subwatershed and practice.
Dissolved Nutrient Loadings
Estimates for dissolved N and P loading for each management alternative
are computed in two steps. First, average annual runoff and percolation are
predicted based on crop curve numbers and soil available water capacities for
each rotation and soil group. These estimates are then combined with
dissolved nutrient concentrations in runoff and percolation based on soil and
crop management characteristics, resulting in average annual nutrient loading
estimates to edge-of-field.
Runoff and percolation estimates for base conditions in subwatershed A
are contained in Table A-26. The average runoff and percolation estimates
(in cm/year) for each rotation in Table A-26 are multiplied by the crop
acreage data for subwatersheds B, C, and D to obtain average annual water
movement for these subwatersheds, as presented in Table A-27. The effect of
varying management practices on water movement in each subwatershed is
estimated in Table A-28, for combinations of contouring and spring plowing or
no-tillage.
Dissolved nutrient concentrations in runoff and percolation are
presented for subwatershed A in Table A-29. The predictive equation coef-
ficients listed in the appropriate table in Appendix III for each crop and
practice are combined with the values of the input parameters given
previously, to determine the average nutrient concentrations for each crop
and rotation in Table A-29 (assuming spring fertilization for all crops).
These same concentrations are combined with the predicted average annual
runoff and percolation for the other subwatersheds to determine the average
annual loadings in Table A-30. Predicted nutrient fluxes under various
management practices are summarized in Tables A-31 and A-32. The ranges for
each nutrient represent the levels of loss expected for spring fertilization
vs. fall fertilization for all crops. The actual loadings should fall near
to or within this range. Dissolved N loadings are usually greater for fall
fertilization, as is expected. However, dissolved P loadings are higher for
spring fertilization than for fall fertilization. This is due to the
reinitialization of soil available P at the beginning of each model year.
Because soil available P levels generally increase over the year due to
fertilization, the spring applied fertilizer will have a longer period over
which it may contribute to runoff losses, whereas fall applied phosphorus may
only contribute for a few months before the soil P levels are reduced by
reinitialization. Phosphorus loadings are rough approximations, and the
decrease in P loadings due to changing the time of fertilization is not
intended to be modelled by the predictive equations. Rather, the ranges in
Table A-31 and A-32 for available P more closely represent the normal range
to be expected for either time of fertilization.
174
-------
O
iTERSHE
=3.
LU
LU
C£.
LU
o
LU
re
1—
1 — 1
00
(— (
h-
oo
1 — 1
LU
h-
2
DC
(~J>
^J
1 — 1
o
oo
oo
CM
c
o r— re
i— -r- 0
CO i —
H-> E
n —t
tn T- o
CO r—
•M E
i— re
CO T- 0
CO i —
E*-t-> E
re r— re
h- o -r- o
i — CO i —
E
re
to o
E
re
LO O
E
re
^ o
.
-M >> E
oo i — re re
•r- I— O
to o i—
CM r— re
•r- O
CO i —
.O "^ £Z
•r^ ^»") cz
i— T- r— O
CO O i —
C"
O
•r-
U
• r-
ll-
CX CO
3 CO
o re
CD O
r— r— ••
•r- •!—
O 0
oo oo
CM
o en
i — r—
CM
IN^
o en
en
LO
Q en
LO
c_> to
CM
° £ r^
to
DQ IO
to
CO to •
r— r— —
OO
LO
0 0
CM i—
LO
0 CM
CM i —
OO
LO
0
O CM r—
„ i
OO
E
o
r~i
0 ^
i~
0) >,
O T-
•r- CO
oi c
O 0)
•— TD
O "— -
S— ^^ -^
•a - i —
nr >=c DQ
i —
to
^t"
to
oo
LO
to
OO
2
to
CM
to
en
•vi-
to
rn
Q.
r—
•n-
o
oo
LO
LO
LO
LO
O
CM
O
CM
0
CM
LO
00
LO
LO
00
>j
re
o
tu
0
i.
tu
a.
0
oo
LO
0
oo
0
oo
0
oo
LO
LO
o
to
o
oo
o
to
— ^
^5
^
eu
+j
ID
E
o
•r—
c.
re
O)
o
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
LO
CM
^ s
O)
^
0>
<* *
°-
o
0
LO
O
o
LO
o
0
LO
o
o
LO
o
o
LO
0
o
LO
o
o
LO
o
0
LO
o
o
LO
o
0
LO
0)
^^
O)
^
0_
tu
re
CL
1
-a
•r-
r—
o
00
00
oo
CM
en
oo
CM
LO
OO
LO
oo
LO
oo
00
en
LO
CO
O)
*^^_
•0}
^^*
ca
en
CM
^
r-1
to
00
<—
tn
oo
r-'
to
en
LO
en
LO
tn
to
CM
tn
•=3-
to
o_
to
CM
oo
r—
CM
CM
CM
to
CM
OO
OO
to
OO
CM
to
*• — ^
re
^
O)
oo
o
s^^ f
•z.
oo
oo
oo
o
to
OO
oo
CM
CM
CO
OO
en
oo
00
oo
^
re
0)
-^
**— *
oo
o
o
LO
*
r—
O
LO
^
O
o
LO
r*
O
0
LO
o
o
LO
n
o
LO
o
LO
o
o
o
**
oo
o
o
LO
ft
o
o
0
oo
0)
-^
a>
^-^
*s^
oo
oo
o
o
LO
o
o
LO
o
o
LO
0
o
LO
o
o
LO
o
o
LO
o
o
LO
o
o
LO
o
o
LO
o
0
LO
01
-^
o>
v~""*
Q
I/
oo
175
-------
Q.
ID
O
o;
CD
i — i
O
C/)
>-
03
^
LU
1 1 1
111
o;
LU
•^
o
^SZ.
o
oo
,
+-> O CU
O 4- C
r— 0
n:
-O
CU
"55
S-Q
CU
4-J
03
-o
CU
.C
I/)
— • i-CJ
O3 CU
-E -i->
O3
1/5
ro
CU
-o
O)
JC
CO
S_CQ
Ol
^0
2
-D
cu
CO
CU
ro
S
c E
••- 0
CO *t— )
O fQ
i- 0
oct:
•r—
O
OO
00 «3- OO COLO i— OOOO OO
^j- oo . co i— ^ CTI LO
CM
oo oo o COLO r— LOI^. LO
CTlCOCOi — OOCOCMCNJ
r— OO CO LO
CM oo r^ coo LO coco co
co oo LO coco co «=J-co r~x
LO i — OOl — •* i — OO CO
ro ro ro ro
co co cocu , S. >, S~ S->>i->>
OO OO OO O OO OO O O03 O03
oco oco (Jco o Oco uco O O-C u^:
i — CM OO «^-LO CO I~^.OO CTl
*d" vl- CM 1 "3-
CO CO C\l
•—i— CO
o3 03 O3
cu to cu cu
.C E JZ JZ.
S 03 S 2
E E -O E C
O O O O 03 O
O O CO O -E O T3
cu
> a.
n
03 1
0 -(-> -Q
r— O 3
1— CO
176
-------
0.
oo
oo
o
1
—1
oo
rs
fV
LJl.
O
— 1—
ml—
D_
oo
O
n:
ex
LU
oo
et
Q-
1
O
i— i
—i
O^^^^
"^^"
OO S-
Q \
i-
E
"•i—
z. oo
— 'OO
o
oo _i
oo
O h-
LU
Z. i—i
LU CC
C!3 I—
0 5
o: z.
h-
i — i •>
z. s.
>,
LU \
oo h-
<
:r:co
Q- 0
Q "~
1-1 z.
| | !
0
oo oo
oo
«o
-—N_I
— i
OO _!
*•— ^* I-H
o
oo oo
oo —
O~,
_l ±£
LU
J 1 1 1
i— i a:
O 0
oo
^~
O LU
LU Z.
\— O
< 1
|_J 5"
^^ ^-«
I— O
oo a:
LU LU
U~l
CN
k
LU
_l
ca
<:
i—
to
H3
CU
O
-(->
O
ta
i.
Q.
CMLOCNjco^-r^ooi—
cr>c\icococo^-oooo
tOlOCOCMOOOOi — i —
cMCM«d-r--«3"cy>oooo
r— LOCOi— r— VDi— O
r^» ^~ CM o*& LO co r^ LO
CM CM r— i— r—
CMLDCMCO-vt-r^-COi—
cncMcocooo-vfooco
VOlOCOCMCOOOi — i —
LD^Dr^.^t-CMI^U3C7l
criCO<*coLn^t-cMi—
UDCVjr^LDtDi — COCO
LOi— r— UDLOOOCMCO
•^- <=J- CM i— CM CM i—
Lr)co-=i-coLnLncor-~Ln
CM CM i— r— i—
i~v.Lrjio^-co«Di-^o
COr— Or^-CMr— LO«sf
CM CM i— i— i—
CO n— CT> CM
CnCO^COLD^t-CvJr—
CMr~-.ooco^t-uDO
to c\j co f^o O") r^« o^ ^s*>*
CO CO ^- r— i— i—
oococni-ocMr^uDcyi
a^oo^i-coLn«^-oji —
CC dL C£ C£
oooooooooooo
r*f*r\t\rtn*\r,
Q-Q-1— h- D-D-h-h-
Luoocc:2:LuooQi2:
cu
co
to
r—
,f_
•)->
o
c
II
1—
z.
cu
CO
(T3
^
•r-
+->
-a
O)
o
3
-a
O)
•a
cu
H ~S-
0
h- "OL
o:
&_
3
O
•4-J
2 c
O 0
i — O
Q.
II
co
c
•i- O
i.
Q.
to
•a
II CU
I
Q- i —
oo a.
2
o
i_
2
O -(->
Q. CO
•r-
i— ra
r- S-
n3 -t->
M- to
II II
a. D;
Lu OO
r3"
177
-------
1
_l
cn t- -r- ^
re o -MOO
s- 4- re E
> Q. O O
•=C ct: i—
^
i.
^j O_ C^j
ro o
'o ^1-
o
CO -^
Q_ i-
Q. ~^
E
(J
f\
^ s
E i.
CU O >-
CD S- T- \
ro o -MOO
i- 4- re E
CU 4->VO
> CC O O
< cc: i—
N^.*.
^_^
b
-<
4->
CL ro
O -I-)
t- 0
0 CC
CL
r— ^3
•r- O
0 i-
00 C3
p_
•
o
, — ,
0 O
1 — 1 —
r~~ r^^
CO CO
, — ,
r- r-
C\J
•
o
CM CM
O O
CO CO
CO OO
CO CO
en en
oo oo
-— n-
, ,__
r~* r~
co en
r— CO
i i
o o
(/)
E
re
cu
E J2
O 0
O OO
r-
r—
.
r—
r- r-
r— •—
00 CX)
r~" i—
o o
en en
<— <—
p^.
.
f-^
r^. r^.
i — i — •
en en
CM CM
IT) LO
CD O
CO OO
CO CO
CO CO
O UD
r— CO
O O
c/)
c~
re
^
E JD
O 0
0 00
CM
OO ^~ '-O r— ^" r^» CO "'J" LO
• •« • •• • |> • |
CD ^— ^~ r— f~^ CD r-~ CO ^"
rooo ^ LO uo r-r- *• ^ oo cocn ^-^ ^-r^
O O i — O O i — i — O O O i — i — CO OO ^J- "vt"
oo t*-. LOUO CMCM co cncM i^cn r~. oo r~~.cn rv.
CMCM OO COCO CMCM CM i— CM CMCM CM CMCM CMCM CM
1010 I*-, coco oo Is*- oocn ^i-oo LT> ^t- ^s- ^j-co ^j~
oo co ovo coco en OCM coo co •— •— ooo oo
CMCM CO COCO CMCM CM CMCM CMCO CM CMCM CMCO CM
LO CO CM ^~ CO CD r—* 00 CO
• •• • •• • (• • i
OOO r— Oi — r— *3"CM
u->LnoocMCM^-«d-ooocnooo oo oo r— to
O O O O O i — i — • O i — O i — i — ' ^t" *^~ CO CM
cncn en r-^r~^ cncn en i^<* cn^o en cncn cnvo o~i
CMCM i — i — CMCMi — CMCMi — i — i — CMCMr — i — i —
vo vo co ^ vo co co ^ "^" ^ vo oo ^j~ r^ r^ vo co vo
OCD en f^-t^. OCD en co«d- CD +-> -t->
ro ro re re ro ro ro
cocu to tocucu cu cu tooi cu
E.E E E-E^= J= -E E -E JE
ros re re22 2 2 re2 2
QJ ^^^ Qj fL) ~~^ ~-^, *^« *^s^ QJ ^^ "^^
C-QEEJDE-QEE E EEJDE E
oo o oo oo o ore ore o oo ore o
ooo o ooo ooo o o:n o:n o ooo o:x o
oo«3-Lnvor^oocn o
CO
uo
re
178
-------
1
1— 1
fc
o
§
(_>
LU
•S
CO
g-^
Cw
0
CO
8
LU
i
ijj
-at
i
S
*?
LU
LU
O
LU
§
X
DC
O
LU
HH
^
LU
I—I
t
LU
Q.
O
Lu
CM
*t-
LU
_l
3
r-
0*0
11
£
at
s
c
o
4-1
ro
I'R 1
§fO* «
•00 o o
CO U
*•— XI J-
«-*- £4
"Sg S2
§'F- flj •*-><«-
r— p 3
fl t ^
C i-
= tg
fc §
> -r-
ro
CO *O
^ C
01 Ol
^z o
I/I
**S
3
•-!
0. ro
F- Q
toooo r^moo ino>
• • «ii i i* *i*t ••
F— OF— F— ro CM into
in o LO F-CMO ooi—
LO O CM CMCMV fOO
CM
OOLOCO* O O^- 00 COO
• . 1 1 . 1 . 1 .
O'-oo F- IOF- vo ocn
F~* ^~ CO
O^^F— CM COcn O tOO>
« • • *| .|. .|*| ••
COCMOO F- r»o «* r^to
CM , £ >, E E^ E >, £
8000000 oooooorooroooooroo
VI UM O IO U Ul/l UVI U U JT Uf U UVI O -C U
t/l
i— CM co^into^^oocft o (O
o
I—
179
-------
A-
^
O
1
^
UJ
UJ
tx
O
UJ
o
IT
1—
1
U_i
>
I—H
1—
^— •
QL
Ll_l
i
1 —
— 1
t—
UJ
LU
cs
r^
^
Q.
O
o:
o
o
u_
z:
o
i
1 —
, J
O
rv1
1 1 i
J-LJ
Q_
Q
QJ
0 C
1— O
n:
Q
T3
CU
r~
to
53
^_>
to
3
O
T3
Q
c~
s2
to
CQ
X3
O)
-C
CO
S-
CD
1o
s
1^^
-o
J=
CO
S-
CL)
^_}
to
2
"^
f
o
o
S-
dl
M-
4—
0
c
J_
•
"o
o
i-
cu
Q.
4-
4-
O
C
3
•
O
O
i-
OJ
Q.
runoff
r—
O
o
s-
CD
a
4-
o
c
3
^
•
O
(J
s-
OJ
Q.
tf«
4-
O
C
3
S-
^
0
o
•r—
+J
0
tO
s-
a.
CM CO r— O O i—
•* O CM r- 00 r—
a> i— i— i— co sr
CO CM CO CO CM CTl
CM co ^~ ^~ r^ co
CT> i — CO CM CM VO
• •••••
CO CTt CT» O *^" ^J"
r— i— i— CM CM CM
t— oo co LD uo o
• •••••
o r-v r^ co CM CM
CM 1— 1— 1— p- r—
o LO co «* co co
CJ> LO CO CO F^~ OO
CO >3- **• •=* LO LO
O") ^J* ^O O^ O f**"*1
• •••••
CO O Ol CO 00 CO
"=f ^ CO CO CM CM
i — LO i — o en o
co oo CTI r-x LO r^
Cx) CM CM CM CO CO
o co CT> oo oo rv.
r— LO ^f co r~. co
CO CM CM CM r- r-
co CM ^J- ^t- LO en
• •••••
LO r-> r^ r~~- o o
r— r- r— r— CM CM
co ^j- CM o en LO
• •••••
^- CM CM CM OO OO
r—* r— r~~ r^*
a: a: a:
OO CO CO C_) O O
•N *\ « «* »1 *»
Q. a. 1— a. a. I—
LI_ oo a: LL. co a:
i ^
c
to
0) r-
C71 CJ> Q.
C fO 4->
0) -i- r- 2 E
C 5 r— O 10
•r- O T- S- i—
2 r- 4-> 1 Q.
o a. -(->
•— -a j= i-
CL Q) CD C7) 3
C O -r- O
i— T- 3 IO •»->
r- S- -0 S- C
IO Q- QJ 4-> O
4- CO i- to O
II II II II II
a. o- i— a:
u_ oo a: c/> cj
tC
180
-------
a
§
u
a
LU
0£
O
in 1-1
UJ I—
C9 N4
Z -H
ae ac
• a.
i
§
in t- .
ui cn
•i- C '
u
• f.
Vt 01
ut
U> 3 •
Ut Cl
5 C
cu «r
Q- O
O
c
in
a.
65
CM
00
^- in co i CM to i CM
CM oo cn mm CM
n CM o
m to
i— m
CM O
CM r-^
rocvjcvj^-
coroin
in
o
co
•a-
OO>— i— OOr— OOf— i— OOOr-r— rOCO
CM CM
o o
in in n CM CNJ
i-^ o o o o d
n o CM no oooo r— vo
i • • •
oooo * •* vo pji— ml- o i~^i~. inr-, ir
roro 10 mm mm 01 cr>oo cor-~ oo ^ tx cnoo en
coo rom ^ r^c>u o< — oo
oromi— r^cvjcocvi m •— «t in r- «*•
CM cxii — cvim CM
•* m ^ m i— m
vo CM mo r— ^ o coco r*. o 0% CM r^ cn f** cri i— r— oo^j- co
cn n mi— i*- CM n CMO «3-1— ^ ^ i— ^-o ^ mr— mf*- m
r~ r™* CO CO CO CO CO CO CO CO CO CO r— t-~ CO ^x CO CO CO CO ^^ CO
r-^ i-^ coco 0003 m mm coco oo r-^ cn oo ID co cocci co 10 co
10
ai
u> in co u>
~ J= c
^ s
S i
10 , £ >> E £ •§, E >, £
35 35 35 3 35 35 3 35 35 3 35 35 3
181
-------
-
Lu
3;
cr
c
U_
^
o
I—
§
h-
CO
LU
CO
to
o
_]
J—
z
LU
O£
1—
ID
z
O
CO
<:
LU
_J
CO
?
o
1
w
L.
•
io.'S
3 -*
C£ -~-
.
*o IT
uz >,
t- ^-
CO 1—
a. —
i»- -— •
t- t.
o z >,
C ~v
3 1—
a -—
<4- S.
t- X
o •--,
c a. a
3 ^
a •—
" ^^ ^
o s-
u >,
I- Z -v.
0) 1—
a. ~—
it- ^~N
M- S.
O >i
c z ^
3 1—
ee —
CO
•o
Ol
.c
10
(.
01
+J
S
<*- t.
ti ^
o -C
§0. a
-*
a: ^-«
•
"o IT
i-z <;
0) 1—
Q. —
t- !-
0 >,
C Z ~^
3 I—
a: *—
C C
•r- O
10 +->
Q. fd
£+J
O
u a:
•"^ 3
o o
5 °
o
$
co
r— •
^_
in
CO
CM
CT>
r>.
in
o\
r~.
«*
CM
in
co
^~
to
S
Ef,
35
i-
•* !fi ~
oo ii i i en
r-
r— CO O
00 II 1 1 to
«• 1— l-~
o CM ii i i en
in <• in i in i oo
•— i— to r~
i— to
VO <• CM 1 IO l<±
* i— i— CM i—
CM
O CM CO «t r-
00 CM O CO 1 •»
CO
CM in •— o ^ i— in
in to in
CO
»— in n- o r— en CM
CM * i—
r-~
t-* 00 ft O CO CO to
co o o o o co r-.
^~
•<-> •<->•!->
Id Id Id
VI to Ol IO to CD 4)
c c -c c c x: .c
•d id X id a X X
at 0) ~^ 01 ai ~x. ~^
If 1"! 1 If If 1 Is
<_> 00 O OO O O OO O OO O OX
CM co «i- in to r-» oo
f^ 1 00
I— CM
CM to
00 1 00
CO CO
to 1 CM
r— 1 CM
p~ r~-
r^
*—
CO 1 r—
CO CO
in
l~» 1-
CM i in
i^.
CM P~ •!»•
oo o
r>.
•—
CM CO 00
tO i—
in
CM O CO
o in CM
r-~
+J +J
id id
„ „ „
Z ^ S
£>> £ £f,
O Id O O O
o :r <_> o oo
o» o
in r~
i to oo
CO CM
^~
o in
i en i—
CM r^
r— O
1 CM 00
CM O!
1 r— CO
O i—
r~- in
CO
i CM in
to en
in w
i- «•
i co r^
•3- CO
i i in
en
CM
CM
i i r^
oo
r-x
•3-
i i
00
o
+•> •«->
Id Id
J ^
X X
^^ "E
O id O
o z o
(/)
Id
1 '
0
1—
c
o
4J
Id
N
^
+J
s-
ai
t-
cn
c
U
a.
•»
2
e
4->
.C
Ol
Id
(.
+J
V)
§"
'a.
r-
Id
1 1
S-
182
-------
*T^"
,^>
•—
z
LU
8
cc.
1—
z
1
LU
1 1 1
O
1 1 1
g
^
r~
^^
.___
UJ
UJ
en
1
Q_
O
o
Qf
Q
^
1
1
00
ll J
I/)
O
_J
1—
21
UJ
o;
I—
z
r—
ro
^.
UJ
CO
rf
?
•2
o
K-
Q
•a
•c
CO
1
u
T3
1
CO
ro
3E
CO
ai
V)
S-
cu
(0
«£
TJ
O
Ol
s
00
CO
LO
1
CO
•*
o
in
co
i
0
CM
r—
ol
Ol
§
1
o
o
in
CM
CO
i
CO
o
CO
1
0
Ol
01
CO
r^
\
co
0
8
1
CM
to
1
^f
oo
Q_
Lu
o
1
0
in
CO
oo
o
^f
i
CO
o
0
o
01
1
^
oS
co
o
Ol
i
o
10
LO
„.
p^
f*-.
1
00
o
CM
O
0
§
01
1
Ol
01
o
10
1
o
0
00
10
^f
a_
oo
o
o
in
§
X
CO
o
1
Ol
oo
o
o
Ol
1
o
•^
1 —
oo
i
§
o
CM
Ol
1
O
00
LO
^
fX,
,__
1
IT)
8
0
1
o
in
co
Ol
ro
co
0
vo
i
o
in
CO
i
0
ce
oo
cz
o
00
5
i
o
vo
n
o
CO
^f
1
in
00
o
oo
Ol
i
§
in
CO
10
vo
o
oo
Ol
o
CM
^
f^
^^
i
CM
O
CO
Ol
1
0
§
1*^
f—
1
00
o
1
o
LO
in
i
oo
CJ
a.
LU
O
LO
01
1
O
0
CM
*SJ-
CM
1
VO
CO
O
o
1
CO
CO
LO
A
LO
g
CM
1
O
CM
tv.
CM
O
1
^2
CO
r—
o
CM
Ol
^^
LO
1
CO
o
Ol
vo
t
o
•a-
LO
Ol
INI
1
CO
Q-
00
O
8
LO
1
O
vo
CM
^f
CM
1
CM
Ol
O
O
1
8
CO
LO
1
OO
o
CO
o
s
CO
o
1
CO
0
^—
0
Ol
f1^
LO
1
o
0
1
o
in
in
Ol
CO
CO
CM
O
ce
» 1
cn 01 Q.
c: ra
cn •- ^- s S
C 5 i— O O
§'"" ,2 +S ^" "o.
Q. +J
CL Ol CO Ol 3
C (J -r- O
I— -r- 3 tO •!->
i— V- T3 i. C
ro Q. o
VH t/t i- to u
n ll n it n
o. CL. i— o:
LU oo o: oo cj
fO
183
-------
oil
oo
o:
o
:r
O_
oo
o
Q_
LU
LlJ
>-
I-U
z
o
oo
CD
<
<
D.
O
o;
•o
o>
to
S-
cu
oo
oo
oo
o
CN
p-> I
CQI
1 —
ro
•!->
O
1—
Q
O
CD
<
n3
CU
O
•i —
•4->
O
re
S-
Q-
0
00
CM
to
t
O
CM
^—
CO
o
LO
o->
1
o
CTl
CM
O
CTl
to
CM
1
O
r—
LO
OO
o
LO
CO
p"^
1
o
o
oo
CM
o
CTl
r-v
o
CM
o
C£
oo
r,
D_
U_
O
OO
OO
LO
f
0
LO
CM
l~^
O
«=f
00
1
0
cr>
•~
o
CTl
CM
CM
1
O
LO
^-*
OO
o
o
LO
r—
1
O
00
CTl
O
CTl
to
1
O
CM
CTl
a;
oo
ft
Q-
00
0
CM
LO
1
0
to
CTl
to
0
**
00
|
O
r~-
1
o
^t
CM
CM
1
O
OO
o
oo
o
LO
«tf-
I—1
I
o
r^.
CO
O
r-^
to
I
O
CTl
OO
C£
oo
•\
1—
o;
o
*d-
00
LO
i
o
LO
LO
to
o
r-.
i~^
i
o
CTl
CTl
O
to
CM
CM
1
O
o
CO
CM
0
oo
to
r—
1
o
«3-
CTl
O
00
to
1
0
OO
00
(_>
t*
o_
LJ_
o
to
r~v
OO
i
o
CTl
o
LO
o
o
to
o
00
CO
o
o
to
1
o
CTl
^_
CM
o
to
o
p-_
1
o
5-
o
o
LO
1
0
l^s
10
o
ft
a.
oo
o
to
LO
oo
1
o
00
to
^~
o
CO
LO
1
o
CO
r»».
o
CM
LO
r~
1
o
o
o
CM
o
00
CTl
1
o
*3-
CM
0
CO
^~
1
o
CM
to
o
ft
t—
01
2
o
CU r—
CD Ol Q.
C ro
CT -r- 1— ^ S-
C g — 0 0
•1— O T- i- 1
•g i— 4-> t Q-
O Q. 4->
r- T3 -C S-
O. CD CU CD 3
C O -r- O
i— T- 3 n> -M
r- S- -V S- C.
(C Q. O) +J O
M- to S- to CJ
II II II II II
a. a. \— cc
u- oo cc: oo c_>
^^^^
ro
184
-------
REFERENCES
Barnett, A.P. 1977. A Decade of K-factor Evaluation in the Southeast. _In_:
Soil Erosion: Prediction and Control. Soil Conservation Society of
America. Ankeny, Iowa. 97-104.
Beaton, J.D. and S.L. Tisdale. 1969. Potential Plant Nutrient Consumption
in North America. The Sulphur Institute, Washington, D.C.
Brady, N.C. 1974. The Nature and Property of Soils (8th ed.) MacMillan, New
York.
Bureau of Reclamation. 1978. Drainage Manual. U.S. Department of Interior,
Washington, D.C.
Enfield, C.G. and B.E. Bledsoe. 1975. Kinetic Model for Orthosphosphate
Reactions in Mineral Soils. EPA-600/2-75-002. U.S. Environmental
Protection Agency, Corvallis, Oregon.
Haith, D.A. 1973. Optimal Control of Nitrogen Losses from Land Disposal
Areas. Journal of the Environmental Engineering Division, American
Society of Civil Engineers. 99 (EE6): 923-927.
Haith, D.A., A. Koenig and D.P. Loucks. 1977. Preliminary Design of Waste-
water Land Application Systems. Journal of the Water Pollution Control
Federation 49(12): 2371-2379.
Haith, D.A. 1979. Effects of Soil and Water Conservation Practices on Edge-
of-Field Nutrient Losses. In D.A. Haith and R.C. Loehr, editors.
Effectiveness of Soil and Water Conservation Practices for Pollution
Control. EPA-600/3-79-106. U.S. Environmental Protection Agency,
Athens, Ga. 72-105.
Hamon, W.R. 1961. Estimating Potential Evapotranspiration. Journal of the
the Hydraulics Division, American Society of Civil Engineers. 87(HY3):
107-120.
Hanway, J.J. 1962. Corn Growth and Consumption in Relation to Soil
Fertility. Agronomy Journal 54(2): 145-148.
Hershfield, D.M. 1970. A Comparison of Conditional and Unconditional
Probabilities for Wet- and Dry-Day Sequences. Journal of Applied
Meterology. 9: 825-827.
185
-------
Hill, L.D. (ed.). 1976. World Soybean Research: Proceedings of the World
Soybean Research Conference, Champaign, 111., 1975, Interstate Printers
and Publishers, Danville, 111.
Jones, G.D. and P.J. Zwerman. 1972. Rates and Timing of Nitrogen
Fertilization in Relation to Nitrate-Nitrogen Outputs and Concentrations
in the Water from Interceptor Tile Drains. Search 2(6): College of
Agriculture and Life Sciences, Cornell University, Ithaca, N.Y.
Klausner, S.D., P.J. Zwerman, and D.R. Coote. 1976a. Design Parameters for
the Land Application of Dairy Manure. EPA-600/2-76-187. U.S. Envrion-
mental Protection Agency, Athens, Ga.
Klausner, S.D., P.J. Zwerman and D.F. Ellis. 1976b. Nitrogen and Phosphorus
Losses from Winter Disposal of Dairy Manure. Journal of Environmental
Quality. 5(1): 47-49.
Langdale, G.W., R.A. Leonard, W.G. Fleming and W.A. Jackson. 1979. Nitrogen
and Chloride Movement in Small Upland Piedmont Watersheds. Journal of
Environmental Quality. 8(1): 49-57.
Lauer, D.A., D.R. Bouldin and S.D. Klausner. 1976. Ammonia Volatilization
from Dairy Manure Spread on the Soil Surface. Journal of Environmental
Quality. 5(2): 134-141.
Martin, J.W. and W.H. Leonard. 1967. Principles of Field Crop Production.
MacMillan Co., New York.
McElroy, A.D., S.Y. Chiu, J.W. Nebgen, A. Aletti and F.W. Bennett. 1976.
Loading Functions for Assessment of Water Pollution from Nonpoint
Sources. EPA-600/2-76-151. U.S. Environmental Protection Agency,
Washington, D.C.
Mockus, V. 1972. Estimation of Direct Runoff from Storm Rainfall. National
Engineering Handbook, Section 4, Hydrology. U.S. Soil Conservation
Service, Washington, D.C.
National Oceanic and Atmospheric Administration. 1974. Climates of the
States, Vol. I and II. Water Information Center, Port Washington, N.Y.
Ogrosky, H.O. and V. Mockus. 1964. Hydrology of Agricultural Lands. In:
V.T. Chow (editor). Handbook of Applied Hydrology. McGraw-Hill, New
York, Chapter 21.
Smith, C.N., R.A. Leonard, G.W. Langdale and G.W. Bailey. 1978. Transport
of Agricultural Chemicals from Small Upland Piedmont Watersheds. EPA-
600/3-78-056. U.S. Environmental Protection Agency, Athens, Ga.
Smith, E.E., E.A. Lang, G.L. Casler, and R.W. Hexem. 1979. Cost-Effective-
ness of Soil and Water Conservation Practices for Improvement of Water
186
-------
Quality In: Haith, D.A. and R.C. Loehr (eds.) Effectiveness of Soil
and Water Conservation Practices for Pollution Control, EPA-600/3-79-
106. U.S. Environmental Protection Agency, Athens, Ga.
Soil Conservation Service. 1968. Soil Survey Clarke and Oconee Counties,
Georgia, U.S. Department of Agriculture, Washington, D.C.
Soil Conservation Service. 1971. Soil Survey, Cayuga County, New York.
U.S. Department of Agriculture, Washington, D.C.
Soil Conservation Service. 1972. National Engineering Handbook: Section 4,
Hydrology. U.S. Government Printing Office, Washington, D.C.
Soil Conservation Service. 1975a. An Inventory of Ohio Soils-Crawford
County. U.S. Department of Agriculture and the Ohio Department of
Natural Resources, Columbus, Ohio.
Soil Conservation Service. 1975b. Soil Taxonomy. Agriculture Handbook No.
436, U.S. Government Printing Office, Washington, D.C.
Stewart, B.A., D.A. Woolhiser, W.H. Wischmeier, J.H. Caro and M.H. Frere.
1976. Control of Water Pollution from Cropland. Vol. II. Appendix A.
EPA-600/2-75-026b, U.S. Environmental Protection Agency, Washington,
D.C.
Tubbs, L.J. and D.A. Haith. 1977. Simulation of Nutrient Losses from Crop-
lands. ASAE Paper 77-2502. American Society of Agricultural Engineers
St. Joseph, Michigan.
U.S. Department of Agriculture. 1972. Usual Planting and Harvest Dates.
Agriculture Handbook No. 283. U.S. Government Printing Office,
Washington, D.C.
U.S. Department of Agriculture. 1977. Agricultural Statistics. U.S.
Government Printing Office. Washington, D.C.
Walter, M.F., R.D. Black and P.J. Zwerman. 1979. Tile Flow Response in a
Layered Soil. Transactions of the American Society of Agricultural
Engineers. 22(3): 577-581.
Williams, J.R. 1975. Sediment-Yield Prediction with Universal Equation
Using Runoff Energy Factor. In: Present and Prospective Technology for
Predicting Sediment Yields and Sources. U.S. Department of Agriculture,
Agricultural Research Service, Washington, D.C. 244-252.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting Rainfall Erosion Losses,
A Guide to Conservation Planning. Handbook No. 537. U.S. Department of
Agriculture, Washington, D.C.
187
-------
APPENDIX I
MATHEMATICAL DESCRIPTION AND VALIDATION OF
THE CORNELL NUTRIENT SIMULATION MODEL
INTRODUCTION
The Cornell Nutrient Simulation (CNS) model is a mathematical model
which can be used to estimate edge-of-field nutrient losses from croplands.
Earlier versions of the model have been described by Tubbs and Haith (1977)
and Haith (1979). The current version, as used in this study, includes
several major changes from its predecessors:
1. Water and nutrient budgets are computed for two soil layers (0-10
cm, 10-30 cm depths)
2. Snowmelt is computed by a simple degree-day method.
3. The detention parameter for the U.S. Soil Conservation Service
Runoff Equation is adjusted on a continuous basis according to soil
moisture.
4. Soil evaporation and plant transpiration are computed separately.
5. William's (1975) modification of the Universal Soil Loss Equation is
used to compute sediment losses.
6. Concentrations of dissolved nutrients in runoff are assumed to be
determined by the available nutrients in the surface cm of soil.
The pverall effect of these and other minor changes have been to make
the CNS model less empirical. The overall structure remains similar to
earlier versions, however. Soil moisture balances are computed with a daily
time step and nutrient budgets are computed at monthly intervals. The model
is deterministic; i.e., it is based on readily available soil, crop and
meteorologic data and contains no calibration parameters. This appendix
describes the mathematical details of the CNS model and summarizes the
results of validation studies for Watkinsville, Ga. and Aurora, N.Y.
SOIL WATER BALANCE
The soil water balance component of the CNS model provides daily
estimates of direct runoff (surface and subsurface), erosion and percolation
from the surface (0-10 cm) soil layer and percolation from the subsurface
(10-30 cm) layer. The general structure of the mositure balance is shown in
Figure (A)I-l. The soil is assumed to drain to field capacity in one day and
hence the CNS model is limited to well-drained soils without near-surface
water tables or impermeable layers. The general water mass balances are
188
-------
Figure 1-1. CNS soil moisture model.
RAIN
PRECIPITATION
SNOW
ACCUMULATION
SNOWMELT
EVAPORATION
EVAPOTRANS-
PIRATION
TRANSPIRATION
EROSION
SOIL MOISTURE
0-10 cm
TRANSPIRATION
t^DIRECT
RUNOFF
SHALLOW
PERCOLATION
SOIL MOISTURE
I 0 - 30 cm
DEEP
PERCOLATION
189
-------
Surface layer (0-10 cm):
dlel,t+l = d16lt + Rt + Mt - Qt - EU - Plt (A.I)
Subsurface layer (10-30 cm):
d292>t+1 = d2e2t + PH - E2t - P2t (A. 2)
where Rt = Rainfall, day t (cm)
Mt = Snowmelt, day t (cm)
Qt = Runoff, day t (cm)
Ejt = Evapotranspiration, layer j, day t (cm)
Pjt = Percolation, layer j, day t (cm)
6jt = Available soil moisture, layer j, at beginning of day t
(cm/cm)
The available soil mositure cannot exceed 9j, the available water
capacity (field capacity minus wilting point) of layer j (cm/cm). Rainfall
Rt is assumed equal to recorded precipitation during day t provided the mean
temperature Tt (°C) exceeds zero. Otherwise, the precipitation is considered
to be snowfall. The soil layer depths dj are d] = 10 cm and d2 = 20 cm.
Snowmelt
Snow accumulation (cm of water) is modelled as
St+1 = St + ASt - Mt (A. 3)
where St is the snow on the ground at the beginning of day (cm) and ASt 1S
the new snowfall (cm) during the day. Snowmelt is computed using the degree-
day equation from Stewart et_ aj_. (1976):
Mt = (St, 0.45Tt), Tt > 0 (A. 4)
Evapotranspiration
Evapotranspiration (ET) consists of three sources, EVt, evaporation from
from the soil surface on day t (cm) and TR]t and TR2t> plant transpiration
or water demand from the surface and subsurface layers on day t (cm). Thus
Elt = EVt + TU (A. 5)
and since there is no evaporation from the subsurface layer,
E2t = T2t (A. 6)
Both evaporation and transpiration are based on Ramon's (1961) formula
190
-------
for potential evapotranspiration on day t, PEt(cm).
0.021 D?
PE = — e (A.7)
t T + 273 st
where Dt - daylight hours during day t
est = saturation vapor pressure, day t (mb)
Evaporation and transpiration are weighted by CPt, the fraction of soil
surface covered by crop canopy on day t. Evaporation is assumed to be a
linear function of soil moisture:
EVt = Min [(eu/9!) (l-CPt) PEt; di8lt] (A. 8)
Transpiration from the surface layer is equal to PEt, provided sufficient
soil water is present.
Tlt = Min (CPtPEtjd^t-EVt) (A. 9)
T2t = Min (CPt PEt - Tlt;d2e2t) (A. 10)
Runoff
Direct runoff is computed by the U.S. Soil Conservation Service's Curve
Number Equation (Mockus, 1972; Ogrosky and Mockus, 1964):
2
(Rt + M+ - 0.2 Wt) ,
Qt= Rt + 0.8Ht (A-U)
The detention parameter Wt (cm) in equation A. 11 is determined from a curve
number
Curve numbers are usually selected from one of three values CNj, CNjj,
corresponding to three different antecedent precipitation conditions. How
ever, in the CNS model cold- weather detention parameters are always based
on the wettest antecedent moisture conditions; i.e.,
CNt = CNju, for Tt < 0 or Mt > 0 (A. 13)
During other periods, curve numbers are selected as continuous functions of
soil moisture in the surface 10 cm by the method shown in Figure (A) 1-2.
This method is based on curves developed by the Bureau of Reclamation (1978)
which assign CNj and CNjji to soil moisture wilting point and field
191
-------
Figure 1-2. Curve number selection as a function of soil moisture.
z
o
DC
U
CO
z
UJ
0.56,
SOIL MOISTURE IN SURFACE SOIL LAYER
9 (cm/cm)
192
-------
capacity, respectively and CNjj to the midpoint between wilting point and
field capacity. The function in Figure (A)I-2 is given by
CNj; + eu (CNn - CNjVO.S 9"! , eu < 0.5 "§"1
CNt = (A. 14)
CNn + (9lt - 0.5^) (CNm - CNII)/0.591, 9lt >_ O-
Percolation
Percolation is computed as the excess soil water above field capacity:
Plt = Max [d19lt + Rt + Mt - Elt - Qt - d^i; 0] (A. 15)
P2t = Max [d2e2t + Pit - E2t - d2*2'> °J (A-16)
Sediment Loss
The modified Universal Soil Loss Equation proposed by Williams (1975) is
used to estimate edge-of-field sediment loss X-^ (T/ha) due to rainfall
erosion.
0.56
= (> K(LS)C P
where K, (LS), C^ and P are the standard soil credibility, topographic, cover
and supporting practice factors (Wischmeier and Smith, 1978), A is the field
area (ha), Vt is runoff volume (m3 ) given by 100 AQt, and qt is peak runoff
(m3/sec).
Peak runoff can be estimated by assuming a runoff hydrograph as shown in
Figure (A)I-3. In the figure D^ is the rainstorm duration (hr), Dc is the
time of concentration (hr) and Da is the duration of initial abstraction
(hr); i.e. the time from the start of rainfall until runoff begins. Since
total runoff is equal to the area under the hydrograph,
Q t = qt' DC + qt' (Dt - Da - DC) = Qt' (Dt - Da) (A. 18)
where q is peak runoff in units of cm/hr (q = 0.028 q ' A). To determine
t t t
Da, it is assumed that the duration of abstraction is proportional to the
amount of abstraction (0.2W^ in equation A. 11). Hence Da/Dj- = 0.2W^/R^-, and
equation A. 19 can be re-arranged to give
,t - 0.028A (Jt) (Rt .Qt^.,Ut) (A. 19)
193
-------
o
Q
UJ
o
Q
I/)
cu
03
-------
Equation A. 19 weights average rainfall intensity (Rt/Tt) by the ratio of
runoff to precipitation excess which is available for runoff.
Canopy Development
Evapotranspiration and sediment loss computations require estimates of
CPt, the fraction of the soil surface covered by crop canopy. In the case
sediment, the cover factor C^ is a direct function of canopy (Wischmeier and
Smith, 1978). If L-t is the fraction of the time period between crop
emergence and full canopy which is associated with day t, the canopy factor
is estimated as
CPt ' r
This function is shown in Figure (A) 1-4.
NUTRIENT MODEL
The nutrient component of the CNS model has a monthly time step. The
monthly runoff, percolation and sediment loss values required for nutrient
mass balances are obtained by summing the daily values predicted by the water
balance component. The nutrient balance model estimates monthly losses of
dissolved and solid-phase nitrogen (N) in runoff, dissolved N in percolation,
dissolved phophorus (P) in runoff and solid-phase P in runoff. It is assumed
that all dissolved nutrients are in the inorganic form and that solid-phase N
is organic. Losses of solid-phase nutrients are assumed to be fixed or
adsorbed to soil particles.
Nitrogen
Characteristics of the CNS model's soil N computations are shown in
Figure A-5. Separate inventories of inorganic N are maintained in the top
and bottom soil layers, while an organic N inventory is computed for the
surface layer only. Nitrogen in either layer is assumed to be perfectly
mixed. Only inorganic N in the top cm of the surface layer is considered
available for runoff loss. The model neglects denitrification and other
volatilization losses. Since nitrification is rapid in well-drained soils,
and the model time step is large (1 mo), the nitrification step is not
modelled. Nitrogen fixation is not included explicity. When legumes are
modelled, it is assumed that the plants will scavenge the soil for inorganic
N and fix their remaining needs. Fixed N is not considered available for
loss in runoff or percolation as inorganic N.
The inorganic N balances are given by
Iln + pNn + MIn + RNn + mnOn - UN1n - QNn - PN1n (A. 21)
and I2in+1 = I2n + PNm - UN2n - PN2n (A. 22)
195
-------
Figure 1-4. Canopy growth function.
1.0 ••
o
<
u.
o:
3
V)
<
o
u.
o
0.5
>•
CD
o
<
K
U.
0.
u
tr
UJ
>
O
o
0.5 1.0
Lfl FRACTION OF TIME TO FULL CANOPY
196
-------
N IN PRECIPITATION
FERTILIZER N
SOIL N ,
0 - 10 cm
mineral! -
DISSOLVED N
IN RUNOFF
SOLID-PHASE
IN RUNOFF
N
PLANT
RESIDUES
DISSOLVED N IN
SHALLOW PERCOLATION
INORGANIC N ,
0-30 Cm
DISSOLVED N IN
DEEP PERCOLATION
Figure 1-5. CNS nitrogen model.
197
-------
In these equations,
Ijn = soil inorganic N in layer j at beginning of month n (kg/ha)
On = soil organic N in layer 1 at beginning of month n (kg/ha)
FNn = fertilizer N during month n (kg/ha)
MIn = inorganic N from manure or other organic residues during
month n (kg/ha)
RNn = inorganic N in precipitation during month n (kg/ha)
mn = fraction of soil organic N mineralized during month n
UNjn = crop N uptake from layer j during month n (kg/ha)
QNn = dissolved inorganic N in runoff during month n (kg/ha)
PNjn = dissolved inorganic N in percolation from layer j during
month n (kg/ha)
The organic N balance is
On+1 = On(l - mn) + M0n - X0n (A. 23)
Where
M0n = addition of stabilized (humus-like) manure or plant residue
organic N during month n (kg/ha)
X0n = solid-phase organic N in runoff during month n (kg/ha)
The parameters FNn, MIn, RNn and M0n are model input values. The para-
meter MIn includes not only the inorganic N in manure applied during month n
but also the readily degraded manure organic N which has mineralized during
the month. The remaining model parameters are computed by a series of
submodels.
Mineralization
The mineralization rate, mn is assumed temperature- limited, and can be
modelled by the Van't Hoff-Arrhenius relationship (Haith, 1973). Within the
typical range of soil temperatures (0-20°C), a linear approximation is
possible (Haith et__al_., 1977):
mn = moV!Tn for Tnl 0 (A-24)
where m0 = yearly mineralization rate as a fraction of the average
soil organic nitrogen content
Tn = average air temperature in month n (°C)
Equation A. 24 apportions the yearly mineralization to the various months on
the basis of degree-months. Mineralization is assumed zero during any month
in which the average air temperature is less than or equal to zero.
Crop Uptake
Total uptake of N in month n, UNn (kg/ha) is approximated by the sigmoid
198
-------
function shown in Figure (A)I-6. If sufficient inorganic N is present in the
top layer, all crop N uptake is assigned to the layer; i.e. UN]n = UNn.
Otherwise, the remaining N requirement, UNn - UN-|n, is taken from the second
layer. If UNn exceeds total inorganic N in the two layers it is assumed that
plant needs will be satisfied from N below the 30 cm depth in the soil
profile. The relevant equations are
UN1n = Min[UNn; Iln + FNn + MIn + RNn + mnOn] (A. 25)
UN2n = Min[UNn - UNln; I2n + PNln] (A. 26)
Runoff Losses
Both runoff and percolation losses of inorganic N are based on Tjn, the
average inorganic N in layer j during month n (kg/ha):
,..„,
Since only the N in the top cm of the surface 10 cm layer is considered
available for runoff loss, the "runoff available" inorganic N is 0.1Tj_n.
The portion which is actually lost is determined by the fraction of available
water Rn + Mn which runs off (Qn), and hence
Runoff losses of solid-phase (organic) N are a function of sediment loss
and are given by
ERfjXnOn
Xn = -___ '' (A pq\
xun lOOOp lA^yj
where
P = bulk density surface (10 cm) soil layer (g/cm )
ERfl = N enrichment ratio
An enrichment ratio of ER^ = 2.5 has been used in all applications of the
model. The average soil organic N is
TTn • °" V"*1 (A.30)
Percolation Losses
Percolation losses of N are computed similarly to runoff losses and are
199
-------
Figure 1-6. Crop nutrient function.
o. 75 4-
emergence
100
maturity
PERCENT OF GROWING SEASON
200
-------
based on the fraction of available water which percolates:
Pin
PNln = - _ Iln (A. 31)
Rn + Mn
(A. 32)
Pin
Computational Sequence
Organic N levels are computed by substituting equations A. 29 and A. 30
into A. 23 and solving for On+i. A comparable procedure is used for inorganic
N. Equations A. 25, A. 27, A. 28 and A. 31 are substituted into equation A. 21
which is solved for \i n+i. Similarly, I2 n+i is determined from equation
A. 22 using equations A. 26 and A. 32. These two steps may give negative Ii>n+i
and/or I2 n+1- When this occurs, the negative values are set to zero and
runof and percolation losses computed. Crop uptakes are then determined from
available inorganic N minus the runoff and percolation losses.
Phosphorus
The soil P model is based on an inventory equation for available P;
i.e., that small portion of total soil P which in principle is available to
plants. Interactions between available and either fixed or organic P are not
considered. The model is shown schematically in Figure (A)I-7. Primary
concern is with P runoff losses and only the surface soil layer is modelled.
The following mass balance applies in principle to the total available P.
However, since most of this P is adsorbed, total available P is approximately
equal to adsorbed P. Thus,
APn+l = APn + FPn + MPn - UPn - QPn - PPn - XPn (A. 33)
in which
APn = available adsorbed soil P in surface soil layer at beginning
of month n (kg/ha)
FPn = fertilizer available P during month n (kg/ha)
MPn = available P from manure during month n (kg/ha)
UPn = crop P uptake during month n (kg/ha)
QPn = dissolved P in runoff during month n (kg/ha)
PPn = dissolved P in percolation during month n (kg/ha)
XPn - adsorbed available P in runoff during month n (kg/ha)
The average adsorbed P in the soil during month n is
TF - W" * (A.34)
201
-------
-a
o
E
to
3
s_
o
-i.
to
o
-------
As with the N model, fertilizer and manure P inputs are input parameters and
the P ouput terms in equation A. 33 are computed.
Crop Uptake
The sigmoid growth function from Figure (A) 1-6 is used to apportion P
uptake over the growing season.
Losses of Dissolved P
The concentration of dissolved P in the soil solution is determined by a
linear equilibrium isotherm:
an = 3dn (A. 35)
in which
an = average concentration of adsorbed available P in the soil
during month n (mg/kg)
dn = average concentration of dissolved available P in the soil
solution during month n (mg/1)
g = P adsorption coefficient
Since the concentration an is APn/p, the concentration of P in the soil
solution is
' <*•*>
The adsorption coefficient is determined by a regression equation (Haith,
1979) based on data from Enfield and Bledsoe (1975). The equaiton has % clay
(%C) and pH as independent variables:
2
3 = 5.1 + 2.2(%C) + 26.4(pH - 6) (A. 37)
Concentration of dissolved P in runoff and percolation are assumed to be
the same as that of the soil solution. However, as with the modelling of
dissolved N losses in runoff, only the P in the top cm_of soil is considered
available for runoff losses. Hence in computing QPn, APn in equation A. 36 is
replaced by 0.lAPn, and the runoff losses converted to kg /ha are
AP
All available P in the surface layer is susceptible to percolation loss and
hence
AP
PPn = O-1 FT Pn <
203
-------
Losses of Solid-Phase P
As indicated in Figure (A)1-7, solid-phase runoff losses of P consist of
both adsorbed and fixed P. Adsorbed losses are
(A.40)
Total solid phase P losses during
yp -
n TUUDp
A P enrichment ratio of ERp = 2.0 is used
month n, XSPn (kg/ha) are
XSPn =
PF)
(A.41)
where
PF = fixed P in surface layer (kg/ha).
Computational Sequence
The simulation calculations are similar to those used for N. Loss terms
are substituted in equation A.33, and APn+i is computed. Runoff and
percolation losses are subsequently determined based on APn. Whenever APn+]
is driven negative it is replaced with APn+i = 0.
VALIDATION STUDIES
The CNS model was tested using data collected in two previous field
studies in Georgia and New York.
Description of Testing Sites
The Georgia sites are two small fields in Watkinsville, Ga., that were
monitored for runoff, sediment and nutrient loss in runoff from May, 1974
through September, 1975. Percolation data was not collected. The two fields
have predominantly well-drained Cecil sandy loam soil. Field P2 is 1.3 ha in
area and had no conservation practices other than cross-slope cropping. The
second field (P4) is slightly larger (1.4 ha), terraced and had a winter
cover crop. More detailed descriptions of these fields and their associated
management practices are given in Smith j^t jj_. (1978) and Langdale et al.
(1979). Sampling and analytical procedures are described in Smith et al.
(1978).
The New York testing sites are six 0.3-ha plots in Aurora, N.Y. from
which runoff, percolation and nutrient loss data were collected from January,
1972 through December, 1973. Sediment data was available, but due to
deposition in interceptor collection ditches it was not considered reliable.
The 6 Aurora fields (A5, A8, A9, A15, A20, A21) are a sub-set of 24 plots to
which manure was applied at rates of 35, 100 or 200 T/ha in either winter,
spring or fall. The denitirification losses which are possible at high
204
-------
manure application rates are not included in the CNS model and hence the 100
and 200 T/ha plots were not used in validations.
The Lima and Kendaia silt loam soils at Aurora are moderately to poorly
drained and are characterized by a relatively impermeable glacial till at 1-m
depth. This produces slow drainage and occasional high water tables which
are not adequately described in the CNS model. However 12 of the Aurora
fields are tile-drained, including 6 of the 35 T/ha plots. These six plots
were assumed to be reasonably consistent with the assumptions of the CNS
model and were hence used for model testing.
The plots were equally divided into "poor" and "good" water management.
The former (A5, A8, A15) were harvested for corn silage, hence removing all
crop residues, while the latter (A9, A20, A21) were harvested for grain with
the residues left on the field. Management practices, sampling and
analytical techniques are described in Klausner et_ aj_. (1976a, 1976b).
Drainage characteristics of the Aurora fields are also discussed by Walter j3t
aj_. (1979).
Model Parameters for Validation Runs
Although the CNS model relies on standardized input data which is in
principle available from secondary sources, the determination of model
parameters often requires interpretation and judgement. For example, the
assignment of curve numbers and crop cover factors is not straightforward
even though this information is readily available in tabular form. Field
conditions seldom correspond exactly with the standard descriptions given for
table entries and hence assumptions and interpolations are often necessary.
The model parameters used in the validation runs are given in Tables (A) 1-1
through (A) 1-5. The following discussion outlines data sources and any
assumptions required to obtain the specific values.
Soil and Field Parameters (Table (A)I-l)
With the exception of 9j, K, Ij0 and m0, all parameters for the
Watkinsville fields (P2 and P4) were taken from Smith et_ _al_. (1978). The
available water capacity 6j was obtained from the soil survey (Soil Conserva-
tion Service, 1968) and soil credibility K was given by Barnett (1977). Bulk
density p is the value at 15 cm, and the (LS) and P factors for the USLE are
computed based on field slopes and slope lengths.
For the Aurora fields p, 6j, %C and pH are all soil survey values (Soil
Conservation Service, 1971). Organic N (00) and available P (AP0) data were
provided by S.D. Klausner, Department of Agronomy, Cornell University. Soil
loss factors were not determined for Aurora since sediment and solid-phase
nutrient losses were not tested for these fields.
Initial values of soil inorganic N (Ij0) were not based on field
measurements. In both locations, the model was run from January 1, and the
soil contains relatively little inorganic N at that time. The values of Ij0
205
-------
00
Q
1 1 1
n:
00
py
LU
•
§
2
o
1 — 1
h-
Q
p-H
^
^>
py_
o
U_
oo
ct
1 1 1
LJLJ
h-
LU
1^
**v
e^
O.
Q
1
LU
I — i
U_
Q
2:
^^
1
i — i
O
OO
1
.
. 1
LU
i
OQ
^£
L—
r~
CM
**
o
CM
<
cr>
*^C
LO
n~
"^
00
•=£
LO
«=c
^J-
0-
CM
CL.
i-
0)
cu
£
(13
S-
ro
O_
o-
C
^
p-
O
p-
*-
t
s
**.
1
(
i
c
e
0 f
3 r-
}• •=;
•
— i—
0 «;
•
— p-
*-
CO
£
(
-s
C
x-
4
•r
-» I
o :
- c
— T
0 -i
1J P;
— .
X. A
ft
X.
u:
•> r-
- c
u
i- r-
•
- C
u
t r-
- C
^~
*•"»
£
c
•"^
4
•p
C
r
•^ I
e
= C
J
^ s
T> C
-* 4
r
>> :
^
c
O p-
: ^
1J r
3 p-
•r
<: r
- :
3 r
3
n
a c
3
~ r"
3 P-
P™
•)
• U
3 r-
0
O
3 p-
^s
^.
3
_^
>>
j
j
0
D.
O
J
i;
j
o :
s <
t-
u (
3 4
o :
(
(
0 !
> (
O i
f\
D S
- l~
- «
1-
•)
- y
0
0
u
>>
o
J
J
V
J
[j
2-
n
J I
•9 !
1 1 1
1 1 1
1 1 1
s.
• C
3
1 1 1
1 1 1
1 1 1
oo r^.
- CM CM LO
• • • •
3 O 0 0 C
CO r-
0 CM CO ID
-> o o o c
„
r —
i.
OJ
i- >>
O ro
4-J P—
(J
ro •>
S- <4- -Z.
0
4-> O) O
0 0 T-
ro T- C
>, f- 4-> (O
4-> U C7)
•P- 0 ro S-
i— -r- S. 0
•i- -C a. c
_f!) p> »r—
fO ^0 O}
•a s- c r—'-
O O) -r- ro r
S_ o 4-> T- J
O) Q. i- 4-> •»
O O T- I
p— 4J Q. c J
•p- a. .p- ^.
O "3
CO ' — » CO •>
oo o
C » _l •> r-
3_ M — - D. p- 1
^ f~
j c_
p-
3 U
3 U
„
CM
S-
OJ
ro
rt
^^
CJ
c
ro
CD
O
c~
• r-i
^* p— ^*
O ro i
Z T- J
^ 4_) ^
DO *P~
«: c _
^ «P— «•*
«\
O
CM
»— •*
O
CM
CM
O
LO
i —
CM
0
LO
r-s
CM
o
c
3
C
O
co
00
CM
O
00
CM
CM
O
p^
CO
o
0 C
0 ID C
o
o c
o *d- C
i — C
4
i- r
CU !
>^
ro I
i — (
•r
« 4
Z. r
t
<_> -r
•p- p-
E <
(0
CD
1-
O
~*s *• — x
O r— ro
~ (O -C r
•^ -r- "~s^
31 4J CD
^ -r- ^:
-^ c —
•p-
*»
o
0
p^^
1
ai
LO
i
j
3
3
CO
1 i —
O
i co
00
1 CM
0
D O O
• O CO
3 CM
0
3 O O
• t£) ^J-
3 CM
rO
1J -C
_j ^^^
O CD
_ ^.
* — "
5 Q-
-> CU
O i —
M -C
— * — ^ rO
- ro r—
O ^= -r-
!>. ^^ ro
D CD >
r -^ rO
— »^^
^ P—
D_ rO
— •! —
O "O 4->
3 CU •!—
= X C
[Z *P~ 'P-
•o <+-
«*
« »\ Q
o u_ o.
E a. <
206
-------
UJ
^^
^£
H-
O_
ID
O_
O
py
O
Q
^^
^£
oo
i<"
0
1— 1
1—
1 — 1
— 1
o_
Q_
— 10 f^^
re O T- O-
Ol T- ro
>- r— >
<""> ec£
CX
CJ
•r"
rO_Q
cn
s- z:
o
E
l — i
CU
-i*; c.
rO
-P
O.
' — i
O.
0
s- z:
CJ
zr fC-Q
03 ~ 1— >.
f) C^
Q.
e^
(J
•r~
C\
t- ^^
0
E
i— i
•a
cu
-E
CO
i_
CU
4->
ro
3£
i —
CM
^~.
O
1
•=J-
,
co
^- r^
1 CM CM
1 CM r-
^~
ro
-^
cn
\s
J**^
r—
CTl
fv»
1
^ — ^
LO
CO
CO
•^^
O
LO LO
00 O
CO O
r—
CM
0.
co ai
^—
cu c
>> s-
a: o
o
i— LO
i— cn
i
^?
, —
CM
*d- r~-
CM CM
CM r—
r-~
co o
CM
CU E
>> s-
a; o
r— *3~
i— cr»
i
^ — %
LO
CO
CO
LD ^£>
oo r^.
co o
f—
^~
Q_
r^
^«
^j-
00
*— ^
^j-
v.^-*
00
LO
^—
^- LO
00 .—
CM r—
<- LO
CM LO
CM
LO
^~
CO
r —
, — »
LO
^J-
^~
t-~
f^-^^—^
LO IO
CO •—
^— r—
LO LO
•CTl LO
CM
LO
«=c
r-.
r—
*N^"
00
^-^
LO
•^**
o
r*-.
r^—
LO
^
CO
LO
r-^
CM
LO
i~-
co
^>^
. — ^
LO
CM
r—
^-^
LO
O
CM
LO
*3-
CO
OO
«C
l^s.
t-~
^-
CO
x-^
r^
**_*-
CO
CO
1— .
1— (T)
CTi •—
i— i—
i— LO
«3" LO
CM
LO
r—
CO
p^»
, N
CM
CO
co
^~
^•^^"^
CM LO
^j- ,_
r^ r—~
CM LO
CO LO
CM
LO
P—
cC
r*^.
r—
^J"
00
^^^
<^J-
^ -•
co
LO
r—
•^f- LD
LO i—
CM r—
<* LO
CM LO
CM
LO
r—
CO
r^.
, — «
LO
CM
CO
r~«
^~N.^-^
LO LO
CO i—
i^ r-~
LO LO
CO LO
CM
a-i
r^» r^*
r— r~~
^" ^J"
00 00
* — x ^~^
LO 1—
*^~* *-~*
LD LO
LO «tf-
i — r—
LO <— LO
LO LO i —
CO r— r—
LO i — ID
r~ LO LO
CM CM
LO LO
i — i —
co co
r^ f"^
, — ^ ^^»
LO CM
CM CM
^f O
^•^ *~^^^*
LO CM LO
O CO i —
CM i — i —
LD CM LO
CO «d- LO
CO CM
O i—
CM CM
s_
cu
"O
E
ro
E
S-
t-
o
(4—
co
r^»
CTl
i —
»
o_
~^3
E
ro
CM
O_
S-
o
t-
LO
^^
O1
"~
• •
CM
s_
ro
CU
>-
• ^
&_
CU
T3
E
• r~
ro
E
CU
S-
S-
0
CM
CTl
^-.
^jT
o_
«\
CM
O_
S-
o
ll
^r~
^
o^
^«
• .
r_
S_
ro
CU
>-
ro -£
E
O
•r—
I '
ro
O
•r-
r-—
JC. Q.
•P CX
C ro
O
E «H
o
cn
c to
il -P
3 E
T3 Q
T3
CU CU
i — fO
a.
O. i/i
ro cu
to
CU CU
S- JZ
3 -P
C C
/rt (1 1
>U UJ
E U
+ o.
i- C
a> -r-
N
•(- tO
r— S-
•r* CU
s- E
cu 5
•**. ^"s.
2 0
207
-------
given in Table (A) l-l are based on values obtained by long-term (10-25 year)
CNS model runs. Hence they are roughly equivalent to long-term average or
steady state values. Annual mineralization rate is the most uncertain
parameter in the CNS model. The values of m0 given in Table (A) l-l are
based on general values given by Brady (1974).
Nutrient Applications and Crop Uptake (Table (A)1-2)
Applications of nutrients were obtained directly from Smith et al.
(1978) or Klausner _e_t _al_. (1976a). At Aurora, based on experiments by Lauer
et_ a/L (1976), it was assumed that 85% of manure ammonia-N would volatilize
shortly after application and hence, the inorganic N applications in Table
(A) 1-2 include only 15% of manure ammonia-N. Available P in manure was
assumed equal to the dissolved P content. Crop nutrient uptakes at Watkins-
ville were estimated from yields given in Smith et_ aj_. (1978) while Aurora
values were obtained directly from Klausner _ejt _al^ (1976a).
The organic N in manure was considered to mineralize much more rapidly
than soil organic N. Based on the soil N balance given in Klausner et a!.
(1976a) for the 35 T/ha plots, a decay series of 50%^25%-10% was determined.
Thus 50% of the manure organic N is mineralized during the first year follow-
ing application, 25% of the remaining organic N is mineralized in the second
year after application, etc. This decay series was combined with monthly
temperature-dependent mineralization rates (equation A.23) to determine the
manure inorganic N inputs MIn in equation A.21.
Cropping Dates, Curve Numbers and Cover Factors (Tables (A) 1-3, (A) 1-4, (A)
1-5)
The cropping sequences given in the tables are based on plowing,
planting and harvesting dates given in Klausner _et_ jal_. (1976a) and Smith et
al. However, the estimated crop emergence, 100% canopy and maturity dates
are guesses based on typical values for the two regions. Cover factors are
linked to canopy development and were taken from Wischmeier and Smith (1978).
The CNS model is very sensitive to curve numbers and attempts were made
to make the selections as objective as possible. In both locations, fallow
curve numbers were used from plowing to 10% canopy. The only fields with a
history of organic matter build-up were A9, A20, and A21 at Aurora and these
were considered to have "good" hydrologic conditions. The remaining 5 fields
were all "poor". The Watkinsville soil is in hydrologic group B, while at
Aurora the groups change from plot to plot. Plots A5 and A9 are predomi-
nantly Kendaia (Group C) and Lima (Group B) is the major soil on A8, A15,
A20, A21 (Jones and Zwerman, 1972). However, tile drainage artificially
changes these groupings, increasing drainage and reducing runoff. Each field
was thus assigned to the next lower runoff group, A for A5 and A9 and B for
A8, A15, 20 and A21.
208
-------
CM
D.
O.
LU
GO
CC
O
O
O
LU
CO
LU
o;
ZD
o
GO
LU
03
D-
o
m
i
LU
—I
CO
S_
O
U
OJ
O
O
£
O) O
O
Q.
O
(T3
CD
n. CD
o w
s-. +->
c_> GO
CD
O_
O
O
CO tQ lOLO^CMCMCMCO
00
O O O O O O O
CO
o
to CM en o o o o
CO CO CM CM CM CM CO
o o o o o o o
CTl
co
oo
i CTl
. r-.
co
co
CTl CTl
i CTt
CTl
o o
o
o o o o o o o
r- in co o o
Ol
o
CU C1J (/)
CD i- CD
i- 3 >
Ol -MS-
£EE ("0 fO
LU s in
0000000
r— LO CO O O
s
o
CD
u
CD
CD
s_
cu
Lf,
-p
CD CO
S- CD
3 >
-M S-
fO fO
CM
CO
LO
S-
o
CJ>
i— i— i— CM
r*"^ 0*1 o"^ ^^~
. co
UD
s-
o
o
209
-------
TABLE 1-4 CROPPING DATES, CURVE NUMBERS AND COVER FACTORS FOR WATERSHED P4
Crop
Rye
Corn
Rye
Corn
Date
1/1/74
2/20/74
3/24/74
4/23/74
5/3/74
7/1/74
9/1/74
9/16/74
11/2/74
12/1/74
1/21/75
4/15/75
4/24/75
5/24/75
7/1/75
9/1/75
9/16/75
Crop
Stage
Mature
Harvest
Plow
Emergence
Mature
Harvest
Emergence
Mature
Harvest
Plow
Emergence
Mature
Harvest
°/
Jo
Canopy
100
100
0
0
0
10
50
80
100
100
0
0
10
50
80
100
100
0
0
0
10
50
80
100
100
0
Curve Number
CNn
72
72
72
86
86
74
74
74
74
74
74
74
72
72
72
72
72
72
86
86
74
74
74
74
74
74
Cover Factor
ct
0.05
0.05
0.46
0.46
0.46
0.42
0.38
0.22
0.22
0.22
0.47
0.47
0.30
0.15
0.05
0.05
0.05
0.36
0.36
0.36
0.32
0.29
0.20
0.20
0.20
0.45
210
-------
00
o
LU
00
CC
LU
I—
2
^£
cc
o
^>
o;
o
Lu
OO
rv
LU
CQ
s:
^
LU
o:
o
Q
j|
OO
1 , 1
1—
<^
Q
crs
•z.
i — i
Q.
Q.
O
C£
(_)
to
1
LU
—1
CQ
2
CM
O
CM
<
i — i
l — i
o un
-<
OJ
-Q
£
^^
CO
CU ,
0.
o
c
ro
O
^
CU
CD
rO
4->
00
CU
4->
ro
Q
^ ^, r>. r^ r*.
to P"» r*^ to to
I**** r^« r**« r^- r^«
to r~« r». to to
oo to to oo co
r-^ r^» r~- CM CM
r~» i-~. t^« CM CM
r^. r--. r-^ r~- 1^
to to to •— •—
CO 00 CO CO CO
o o ooo
r— O
r—
s-^
CM
•=C
O
CM
e^ C
QJ
CD
S S-
O QJ
i— . g
Q_ LU
CM CM
CM r^* r^*
f^Sfc ^^ ^v^. ^^
^ — . co rji r^-
r— CM i— i—
i — LO IO CO
r^ r>. i-v r-^ r^ r^ r--
p^. i^^, p^ r*>» r^^. ^^ ^*
to to to r^ r^s to to
CO CO CO to to CO CO
CM CM r^ r^. r^ CM CM
CM CM r~- r>. r^ CM CM
r». r--. r--. r-v r^ r^ r^
n- UD to UD r- ,—
00 CO 00 00 CO CO CO
o o o o ooo
O i—O
^-,
Lf) CM
en o
4-> eC =a; c:
CU (/) ^— ^ ^— ' CU
S- O) CD
~^ j> ^ ^ y
4-> S- O O O)
ro ro •— •— E
CM CM CM
r*-. i — * r*-* co co co
^^ ^^ ^ — * ^- r^ r^.
r"~ tO tO "^^ ^^^ ^^^
i— CM CM Lf) CM i—
o o o -^. ^Z. \
r— i—i— ^- to 00
1-^ I-- 1-^
to to to
p^ ^^ r^^
to to to
00 CO 00
CM CM 1^
CM CM r^»
i^ t~^ r^
^to
CO 00 00
0 00
o
^~
LD
^
LO CTt O^
i— CM C\J
O O O
211
-------
ValIdation Results
Watkinsville, Ga. Sites
Measured nutrient, water and sediment losses are compared with CNS model
predictions for the 17-mo period May 1974 through September, 1975 in Table
(A) 1-6. Precipitation during this time was 123 cm on field P2 and 97 cm on
field P4. Observed losses were taken from Smith et_ a1_. (1978). Runoff pre-
dictions exceed observations by substantial amounts on both fields, although
errors were smaller on P4. Dissolved N and P are over-predicted by
approximately the same degree as runoff on P2, indicating that errors in
these predictions are more likely due to faulty hydrologic parameters than
serious errors in nutrient balances. Sediment and solid-phase nutrient
predictions are quite reasonable, particularly considering the crude and
somewhat arbitrary nature of model predictive equations and parameters for
these losses.
The most critical problem is in the simulated losses of dissolved P in
runoff. Although the magnitudes of these losses approximate the
observations, the large predicted reduction from P2 to P4 was not seen in
observations. As indicated in Table (A) 1-7, this was the only substantial
difference in losses between the two fields that was not accounted for by the
model. The probable sources of error is the absence of a source term in the
CNS model for leaching of P from plant material during the colder months.
January, February and March accounted for 56% of the observed dissolved P
loss from P4 which had a rye winter cover crop. The comparable figure for
P2, which had no winter plant cover, was 29%,
Aurora, N.Y. Sites
The six New York fields are far from ideal as a basis for model
testing. Not only are the sites artificially drained, but the primary
nutrient sources are manure applications. The CNS model is not well-suited
for either of these characteristics. Nevertheless, the Aurora validation
studies were considered essential since the Georgia simulations provided no
testing of either the percolation or snowmelt portions of the CNS model. The
two-year testing period at Aurora had 189 cm of precipitation, 14% of which
fell in June, 1972, when Hurricane Agnes passed over the sites.
Observed and predicted losses for the six New York fields are shown in
Table (A) 1-8. Observed values were provided by S.D. Klausner. Examination
of percolation observations revealed another problem with these sites.
Percolation was improbably high on two of the fields (A5 and A9) suggesting
that water flows were not independent. For this reason, comparisons of the
mean losses shown in Table (A) 1-8 are more relevant than comparisons of the
separate fields. On this basis, runoff and percolation predictions are
relatively accurate. Dissolved N in runoff is underpredicted, indicating
that more manure N was available for runoff than had been estimated for model
input values. Observed dissolved P losses were nearly an order of magnitude
greater than predictions. The CNS model assumed that manure available P can
be described by the same equilibrium relationships as P in the soil. The
assumption appears to be untenable. The overprediction of dissolved N in
212
-------
h-
^-—~
LU Ln
2: r^
p — i en
Q i—
LU
co "
s-
*• cu
U_ JD
u. E
o cu
"^ -i ^
^-. •+— '
ZD Q.
Qi CU
CO
Q 1
LU ^r
> r~-
a; en
LU i—
CO
CQ «
o >,
(C
31 s
1 »,^ f
r— ^—*
I-H
S CO
LU
00 I—
*^. l-H
o co
i— i
i— <:
C_> l— l
i— i C3
Q C£
LU O
Oi LU
Q- CJ3
_j cc:
LU O
Q Lu
o
s: co
UJ
co co
•z^ co
0 O
_l
u_
O 1—
•Z. LU
O i— i
CO CXL
, . i
»— i t
a: ZD
< ^
Q.
^ Q
O Z
0 <
vC
1
M
UJ
— 1
CQ
<
h-
•o
O)
-(->
O
• r-
•o
cu
a.
i
^^
D.
•a
cu ~a
r- CU
U_ >
S-
CU
to
.a
o
T3
0)
-t->
o
• r^
-c
a;
D.
CM
a.
•o
cu
•r- -o
u_ cu
>
i-
cu
to
JO
o
•
en
ro UD "^ i — i — <£>
• • • • • •
(D i— CM CM O O
CM
•^^
"* — — *-. O "* —
fC JD ^J" "O
r--x en o Ln ro 10
• • • • • •
CD i — CM CO O i —
1 —
l£>
CM Ln LO CM ^f- Ln
o en Ln co o ^j-
•tf-
-^
"~^s *^^ (J **^
03 JD i— -a
o co UD •* ro oo
• • • • • •
co r-> ro en o Ln
CM
c: c
fQ C <*"~^11 **"** £Z ff~~^' *~~~*'
_c -i- ro "z. re -i- re rx re
^ _c: J= -c: .c
^-~ i— -zz — oi-^. D---^. m-^.
E — en to en en to en
O "O-^ to -^ -O.^ rd-^
^ — 4-> O) -C (1) JT
C > Q- > Q.
M- O) i— M- 14- i— M- i *t-
M- E OH- T34- O"4- "04-
O •<— to O -r— O to O T— o
c T3 toe i— c: toe i— E
3 CU -r-3 O3 'r-;3 ors
o: co QQ; coo; act: coo;
c:
o
•r~
+J
3
"o +•>
to c
(D
^_ t—-
C i=
•r- +-) •!-
E "O
2: cu cu
i E to
*3- •!—
DC T3 c:
^
O^COO
s: t— Q- i—
>s^-, ^"-* ^^. *^*^
to JD O -O
213
-------
oo
Q
1
LU
Lu
Z
LU
| ! j
s
1—
LU
CO
oo
^y*
«£_
O
1—
<
C£
<
>
Q
LU
t—
O
Q
LU
C£
D_
Q
Q
LU
a;
LU
oo
CQ
o
Lu
o
z:
o **
00 Q-
o; -a
rf" ( —
*•*. t^
a. re
5»"
O CM
C_3 Q_
P^*
1
1
\~4
LU
1
CQ
<
1—
•o
O)
1 j
o CM oo r-- ^j- CTI
•r- ^j- co un r^ uo
"Oil I i |
a>
c
E o_
O
M-
CU
O^ ^"
c a.
T3
-c o
0 -P
4-> CM
C Q-
O)
O
s-
O)
Q_ T3
O
s- o <=i- ^- oo o
aj oo r^. ^- 10 i—
CO 1 1 1 1 +
_a
o
c:
Z •!-
s: a> Q-
to
-o M- d) <+- -C 4- O)
c > o a. o >
l|_ aj , — c 1C i — «4-
«*- E O3 T33 OM-
O -r- tO DC T- O£ tO O
c "o to i — to c
3 O) T- C O C -i-3
oi oo Q •!- oo -i- Q D;
r^
CO
i
CM
1 — .
1
Q-
a>
to
to <*-
j= 1-
Q- O
1 C
•a 3
•r- a;
r—
O C
00 T-
214
-------
I—
^^
^_
LU
I-H
oi
h-
2:
Q
LU
>»
_J
O
CO
co
z.
e^C
Z.
o
1— i
h-
2
o
CJ>
("y^
LU
U_
U_ CO
o ^
Z. CTi
ID r-
CtL
«
Q S-
LU HI
tt. E
LU CD
CO O
CQ O)
O 0
1
z: CM
H- r-
i—i CTi
co "
"^ >-
O i.
I-H (O
1— 3
c_> c
i — i (O
Q o
LU — -
f^S
Q- CO
LU
—1 1—
LU I-H
Q CO
O
^- \s
co o
z. >-
0
3
LU LU
o z.
Z. C£-
0 0
CO U_
0£ CO
O) (O
•— 'o
o o
to i-
to o>
•r- 0_
Q
c—
., —
=
O-
-O
0)
O
to
to
•r-
Q
c:
"^
-o
a>
|
o
10
to
•r-
(^ ^
C
o
•r—
4_>
03
"o
O
S-
O)
o_
q-
o
c
D;
X — *
ro
\
a>
* — ^
re
j^
s — ^
ai
\x
" —
M-
O
-^
Q£
^
fO
c*
-^
a>
—
ii_
^f—
0
^
o:
.. — *
E
O
**«^
E
o
•
cu
St_
0-
•
to
.a
0
•a
a>
^.
D_
.
to
jQ
O
,
~o
a>
a_
.
to
JQ
O
.
•a
CL)
s-
o_
.
to
JD
O
•
-a
cu
a.
.
to
o
O
-a
f—
cu
•i —
Lu
•—
<^
LO
, —
CM
•
CO
p— ~
to
CM
•
O
CO
.
o
CM
.
r--
o
.
00
CTi
.
«^J-
IO
CTl
.
to
to
f-^
to
,
LO
CO
^J.
•
LO
CM
LO
PX^
o
LO
1
r — ,
,
•=3-
CM
CO
LO
cn
CM
to
o
to
•
CTl
CM
O
•
0
c^
, —
.
1
0
, —
to
CM
CO
CTi
CO
o
CO
LO
to
o
LO
CO
CM
o
CM
•zf.
r-
00
CTi
CM
•
i—
, —
f—
co
0
•
o
LO
cn
.
o
00
*
0
, —
•
LO
CO
CTi
CO
o
LO
00
to
.
CD
CO
00
, — .
CM
=3;
cn
r-1
«^-
r—
«^j-
•
f—
CTi
^~
•
o
1^^
cn
.
0
CM
CO
co
.
r\
CM
f^
cn
CM
cn
cn
CO
,
CO
CM
CM
cn
c
(O
CD
s:
215
-------
percolation is not as serious as it may appear. Measured N losses are based
on tile drainage at 100-cm depths, while predicted values are for percolation
from the top 30 cm of soil. Additional N losses due to plant uptake and
denitrification are likely in the downward movement of N to the 100 cm
depth. Also, this movement is not instantaneous, and substantial amounts of
inorganic N will have remained in the soil profile between 30 and 100 cm.
Validation Summary
The credibility of a mathematical simulation model is largely subjec-
tive. No model is a complete picture of reality. Rather, models are sets of
hypotheses concerning the fundamental aspects of physical and biochemical
phenomena. Given the unavoidable errors in data collection and analysis as
well as the judgement required in estimating model parameters, models cannot
be proven to be correct. Comparison of model predictions with field measure-
ment can however, provide some indication of consistency and
accuracy. Based on these validation studies, the CMS model appears to be a
reasonable means of estimating nutrient losses from croplands. It accounts
for differences in crop, soil and weather characteristics and reflects the
impacts of management practices such as runoff and erosion control and
fertilizer applications. However, the model is not a satisfactory means for
estimating the effects of manure management. Neither is it useful in
comparing dissolved P losses from fields with substantially different plant
covers.
216
-------
APPENDIX II
GENERATION OF DAILY PRECIPITATION AND TEMPERATURE
FOR THE CNS MODEL
The daily precipitation and temperature models are independent lag one
Markov processes, fitted to the meterologic characteristics of the area being
modelled. The precipitation model utilizes a binary distribution to deter-
mine whether precipitation is greater than a threshold value (.025 cm). The
binary distribution is fitted to the average number of days in each month
with rain, and is affected by whether the previous day was dry or wet. The
conditional probability of rain on day t was computed from the unconditional
probability by use of a regression equation presented by Hershfield (1970).
The amount of precipitation on a day is predicted using an exponential
distribution, fitted to best approximate the mean monthly precipitation. The
model was found to slightly underpredict the number of large storms as
compared to historic data.
The temperatures model includes the effects of seasonal and local
variations in average air temperature. Temperature on day t is a function of
the expected mean temperature on day t, the temperature on the preceding day,
and a random noise term. From inspection of historic records, it was found
that the correlation of the previous day's temperature with the current day
is roughly constant over time and location, and that the variation in the
random noise term may be approximated by a function of mean temperature.
Precipitation Model
Define a wet day as Pr£_ .025 cm
(if Prt < .025 cm, the Prt = 0)
then Pn (W) = unconditional probability of rain on any day in month n =
(# days with Prt> .025 cm)/(# days in month)
Pn (D) = unconditional probability of no rain on any day in month
n = 1 - Pn (W)
Pn (D/D) = conditional probability of a dry day following a dry
day in month n = .1718 + .8462 Pn (D)(Hershfield, 1970)
then Pn (W/D) = 1 - Pn (D/D)
and Pn (D/W) = Pn(D) • Pn(W/D)/Pn(W)
(this is due to the equal number of W-D and D-W sequences
in any long record)
Pn (W/W) = 1 - Pn(D/W)
217
-------
Finally, define 6n = average daily precipitation on days with rain in
month n (cm)
= (av. monthly precip.)/(# days with rain)
To generate a value for Pr^ on any day, choose two independent uniform
0-1 variates X^- and Y^; then
if [*t < Pn(W/D) and Prt_i < .025 cm]
or [Xt < Pn(W/W) and Prt_l >_ .025 cm]
then Prt = .025 - 6n£n (Yt)
else Prt = 0
Temperature model
define: T^ = average air temperature on day t (°C)
y-t = expected value of T^ on day t (°C)
Ct = correlation of temperature on day t-1 with temperature on
2 day t (dmless)
a = variance of temperature expected on day t (°C)
V = random normal variable of mean 0 and variance 1; may be
t
generated from uniform 0-1 variates by
Vt - (-2 £n Xt) 1/2 cos(2Tr Yt)
then
Tt = H + Ct (Tt-l -ut-l) + Vt (1 - St2)1/2
By inspection of historic records,
a) Mt varies sinusoidally over the year, and may be described by
nt = T + TD sin (.01745 (Jt - 105))
where T = average yearly temperature (°C)
TD = (average July temperature) - T (°C)
Jt = Julian date of day t
b) ?t ranges from .60 to .70 for most seasons and locations, and is
therefore fixed at .65
c) a-j- varies consistently with y^, such that
218
-------
ot2 = 33.412 - 1.169 yt
from analysis of twenty-five year records at Ames, Iowa, Aurora,
N.Y., and Athens, Ga.
The data necessary to implement the meteorologic model is then average
precipitation and number of days with precipitation by month, and average
summer and annual air temperatures, as given in Section 2.
219
-------
APPENDIX III
RUNOFF, PERCOLATION, AND DISSOLVED NUTRIENT
PREDICTIVE EQUATION COEFFIENTS
The tables in this appendix are arranged first by crop, secondly by
timing of tillage, and lastly by soil hydrologic group. Within each of these
categories, predictive equation coefficients are first presented for runoff
and percolation (a, b, a1, and b1) for each geographic region modelled,
followed by coefficients for the nutrient concentration equations (a-j, b-j,
and c-j) for spring or fall fertilizer applications. The following index may
aid in the quick location of a specific table:
220
-------
4-
o
QJ
o
ro
r—
to
+->
C
O)
•i —
O
q_
4-
O>
0
O
4-
O
CD
i — •
-0
ro
r—
~^j
Ol
r—
O
to
•r-
~o
S-
o
4-
4-
O
CD
E
• i — •
£
•r^
i—
to
4_>
E
OJ
•T—
O
•r^
4-
4-
0)
0
u
E
OJ
l_>
~^
E
i-
0)
N
|^
.71
40
S-
O)
4-
"O
E
ro
4-
4-
0
E
"^
s-
s-
0
o
s_
"a
>^
-c:
^~
•r~
o
CO
^
•4-^
T3
CU
^~
r^
a>
-o
o
E
CL
O
&-
'-J
OJ
1
CD
E
S_
Q.
CO
r—
1
CJ
Q
•>
CJ
f
CQ
n
e3^
OJ
CD
ro
n—
i —
•r—
+J
O
E
r — .
C\J
p—
ro
E
S-
0
CJ
i — oj -=t LO r^ co
i i i i i i i i i i i
CD CD CD CD CD
E E E E E
r- S-r— S-r— i-i— i. r— i. r—
rOQ-rOCLrOQ-rOQ-rOCLro
4— to 4 — to 4^ to 4— to 4— co 4—
O OO CD
^t" f — • i — ' r-~ n—
1 1 1 1 1
CJ CJ CJ CJ CJ
CQ
« CJ CJ <;
CJ ** **
CQ CQ
f\ n
< <
aj ai
CD CD 2
ro ro O
^ i— "CL
•r— "r-
4-3 4-3 r^
^~
O O ro
E E 4-
oj r~-
1 OJ
OJ 1
OJ LO
OJ »OJ
i— CTi "
co i oj
1 CO 1
1 — 1 — 1 —
to
E
ro
O)
^} ^1
ro O
-E tO
OJOOLOCDCOCTli— OJ^LOr^
1 1 1 1 1 1 1 1 1 1 1
CJCJCJCJCJCJCJCJCJCJO
CD CD CD CD CD CD
E E E E E E
i-r-i-r— i-r-i-r-i-r— i,
CLro CLro CLro CLro Q_rO CL
to4- tO4- to4- to4- co4- to
i — ^J~ r^v o oo CD
OO OO OO "3- «d- •*
1 1 1 1 1 1
CJ O CJ CJ CJ CJ
r~\
CQ Q "
CQ O Q " •> O
<
O CD
•— CD
CL ro
CD ^
E -i-
•i- 4->
S-
CL O
to E
ft
OO
i —
1
CD
ft
OJ
i —
.C
4-3
• i —
2
E
0
O
CO
I
CJ
,
(O
4-
r.
OJ
OJ
o
OJ
i
• —
s_
0)
0
CJ
S-
-------
oo
O
CO
CO
CD
o:
o
o
03
X
o
rt rt rt o
in o mo
o
rj
o n N ci
o> TH »,n
CK N 03 n
>o in *r «r
N',n «r o
V ft OO-
n n mr<
I I I I
in N -o -o
0* N N *0
-0-0-0-0
1 1 1 1
n N -ooo
in v on
n n nn
in o in o
0- O 0303
O 0- M-O
-o «r ^ ro
rv n oo-
o- oo rs IIT
IM n n n
i i i i
o- -o n-o
O 0- 03 111
-o in iii iii
i i i i
-< w o
ro ro to ro
mono
03
03 03 "O to
03 0- to O
CO 0- T O
-o iii n n
o- TH o- rs
03 0- CO 03
(III
iii co TH ro
iiii
TH -c O O
0- 0- O O
mono
rs
n
o- o- r* n
o-
-------
O
oo
O
«\
C_5
n
CO
Q-
CO
a;
o
CJ)
CM
i
CQ
<
in
u. t
o
z
ce
n
Q n
SSOLUE
1 (
-H CJ
a
o
CJ
OH
2
o
h- fl-
ro o TH in t>>
rs TH ro ro
TH TH CM ro
iiii
TH CJ CO O
0- O O TH
1 1 1
in CD TH «r
O O TH TH
o o o o
cj fl- rs co
o o o o
ro cj in ro
TH cj ro in
o o o o
•o o in TH
ro in co N
0 0 O TH
o in o- ro
o o o o
IIII
ro ro cj -o
TH fl- CO TH
O O O T-»
IIII
in o in o
o o o o
o o o o
o o o o
ro fl- fl- fl-
o o o o
o o o o
in >0 SD -0
CO CJ TH N
CJ CJ CJ -1
o o o o
is o ** ro
o o o o
IIII
o ro rs iii
IIII
o o o o
is ci fl- -o
o TH cj ro
cj rj cj ci
TH co in TH
ro ro fl- in
iiii
co ro TH o>
fl- in -o -o
0 O O O
0 O O O
TH cj ro fl-
o o o o
fl- fl- fl- ro
o o o o
o o o o
CJ O TH IS
0 0 0 O
1
0 TH TH CJ
o o o o
IIII
~o cj o in
o o o o
o o o o
1 1
in o in o
TH TH CJ
ro
oo o o
o o o o
oo o o
THTH 0 0
o o o o
o o o o
1 t 1
>o o* in TH
co rs rs rs
o o o o
-oo o o
oo o o
IIII
rs o co co
iiii
in N o* ro
oo o TH
cj ci ro o
O ^ N CO
o TH o* ro
CO O TH CJ
iiii
rs cj cj cj
IS C*. TH CJ
i i
ro >o o* ci
O O O TH
o o o o
ocj ro in
oo o o
cj cj cj n
0000
rs TH o co
>o rs >o in
-o fl- n in
o TH cj ro
oo o o
iiii
OO CO O
•»•» ro fl-
00 0 O
fill
mo in o
TH TH CJ
O 0 O O
o o o o
o o o o
ro ro fl- ^
o o o o
o o o o
in *o -o *o
co o rs
CJ -0 CJ O
TH o O 0-
•o rs CD rs
TH O O Ci
o o o o
1 1 1
co in ui ro
i i i i
rs fl- o ci TH
0000
fl- fl- -O CO
o o o o
o o o o
1 1 1
ci iii ro ci
co rs rs rs
iiii
o TH TH ro
TH o* cj ro
111 CD O *0
111 fl- 0- Ci
TH -o o ro
fl- co in ro
iiii
TH -o n rs
o ro rs CN
iiii
fl- CO Ci 0*
O O TH TH
o o o o
TH CJ fl- -0
o o o o
o ^ rs ro
fl fl- yj co
o o o o
o -o c- in
CJ -0 -H CO
TH TH rj cj
•o in a> ~o
0 TH CI T
o o o o
iiii
in ro co ex
in -o co -i
O O O TH
iiii
in o in o
TH TH CI
rs
o o o o
o o o o
O 0 0 O
o o o o
o o o o
iiii
CJ ro n ro
fl- -H O CO
o o o o
TH TH o- ro
CJ CI TH CJ
o o o o
IIII
fl- in is co
o o o o
fl 0- 111 0
TH ro fl lil
ro ro ro m
fl- fO O fl
TH ci ro ro
o in o fl-
iiii
tn ci co CJ
*o rs fs en
ro fl- -o co
o o o o
o o o o
o -H n n
o o o o
co cj in co
-H CJ Cl CI
0000
fl ro
-------
O
o
oo
i
Ill
U- o
o o o o
1 1 1 1
m o in o
o
o oo o
o o o o
o o o o
0000
1 1 I 1
rt CMtO tO
OK O- 0- O
0- 0- 0- O
0 00 rt
-^ oo in OK
T ro to ro
o o o o
0- MO- 03
CO DO- tO
IX N O N
lilt
rx *H in CM
O rf n n
O- 0-
o o oo
o o oo
in -o too-
n
D CM « 10
o o oo
1 1 1 1
in o in o
CM
o o o o
o o o o
o o o o
o o o o
1 1 1 1
0- O- O rf
O O -I rt
rH ri CM rt
CD 03 03 CO
o o o o
o -o rx o
o o o o
1 1 1 1
in o in o
rx
o o o o
O O 0 O
o o o o
o o o o
1 1 1 1
CD in ro *H
-0 -0 -O -0
o o o o
CM n CM o-
to to ro ro
0 O O O
ri -i rx co
fs -0
-------
o
o
CM
I
GO
<
in
IL «r
o
2
^3
it.
W
M
fl-
ea (N
SSOLVI
1 (
Q
O
OS
z
o
:RCOLA
B
OL A
»
Q
Ul
:>
_)
co ra
CO
»-4
a
a>
Hi
u. «r
o
D
&
2 «
M
Z
QPJ
[SSOLVE
U C
o
o
REGION
AU A
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 1
O O rf rH
CM o o rs
rv IS rs -0
o o o o
ts PJ n n
rr in in in
o o o o
»• in oo in
PJ co o- o
If) If) If) -0
i i i i
•r in >o rs
o o o o
•H «t -o ct-
a* ~< n ^>
PI n n n
*o n m CM
rs CD co o-
o- n in N
O iH i-t T-<
1 1 1 1
o- <• •* ~o
* If) -0 -0
1 1 1 1
n t -o fs
o o o o
o o o o
CO * O N
o n n M
o o o o
*-< n n Cii
0 O O O
O M n 0-
n PJ PJ ^
o o o o
If) &• TO N
O O rt r<
o o o o
1 1 1 1
O N * t
PI PJ to «r
oo o o
1 1 1 1
If) O If) O
«H rt (M
0-
o oo o
o oo o
o oo o
o oo o
o o o o
1 1 1
O r-t «-< M
CD -CT PI
~0 -0-0 -0
0000
PJ -M CK 0*
o oo o
O 0- tt- -0
-1 0-fs CO
-1 O O O
1 1 1 1
m -003 0-
o o o o
PI -o o r if) IIT
1 1 1 1
n if)N o-
o oo o
o o o o
CO W O CD
o .-IIN n
o oo o
O OO 0
CO Mlf) 03
n »p> n
o oo o
«r con o-
O O-i -1
o oo o
1 1 1 1
(M ncM m
0 00 0
1 1 1 1
mono
-I-H (••<
o
w
-4 rt O O
o o o o
o o o o
o o o o
o o o o
III)
o> o* o *-<
o> o* rs -o
•0 -0 O •<)
0000
01 co •cm
o o o o
M n iii r-i
&• O' rs -o
0- 0- 0- 0-
IT) N 0- ^
o o o -<
-H -o »H in
0 rt ^ rf
M n M ro
UT 03 O -H
-o -o r^ rs
oo n N -H
o »H -< rj
I i i i
P-J T If) ~0
M M M M
lilt
tn in N o*
O O 0 O
o o o o
N r-i oo «
o ^H »-• n
o o o o
o o o o
^ 03 O 111
n n n M
o o o o
» N rt -O
O O *H ^
O O O O
1 1 1 1
M «T If) Pv
O O O O
1 1 1 1
mono
-< -< n
*H
CN
OO O O
oo o o
oo o o
oo o o
oo o o
nn n o
» in in n
vH v-4 iH ^4
rs o to rs
>o N rs rs
rs o rn n
o <-i -i r<
i i i i
n -o -i -o
to to «r «•
i i i i
n ro O 03 O fO
in in -o -o
i i i i
to T 111 <
o o o o
0 t K O
*Q 03 03 0s
pin n PI
con rs o
rs 03 co o*
co -> «r s
O *H »•< »~l
i i i i
•fl PJ 111 N
T iii iii in
i i i i
to t 111 ^0
O O O 0
o o o o
03 PI 03 r iii iii
0 O O O
o
o oo o
M >0 03 •*
P-I to to >t
o o o o
T r-. -H n
0 0 ~ ~i
o o o o
1 1 1 1
in to ts PI
PI n to •»•
o o o o
1 1 1 1
mono
rfrf PJ
•0
PJ
o o o o
o o o o
o o o o
0000
o o o o
1 1 1 1
0- O -< PJ
O •-* r4 -*
fs N IS ts
o o o o
PI O -0 03
PI PI 1H TH
0 O O O
o ct> PI «r
N >0 to O
m n iii in
1 1 1 1
is M -o n
O -H r< PI
to to o PI
-------
oo
O
to
CO
Qi
O
O
CO
in
u. «r
0
Z
U
<£
ro
CL
Q CM
[SSOLYE
:i c
Q"
o
OQ
z
o
1-1
i «
_i
o
u
CL
u. ifi
Z
TH
z r-i
n
Q
Ul
r>
_j
en at
en
W
Q
at
in
LI. i
o
z
D
CC
ro
t-n
z
a ri
ISSOLV
U
Q
O
REGION
AU f
O O O O
o o o o
o o o o
o o o o
o o o o
o o o o
~o rs rs rs
o o o o
o o o o
ro n o o
o o o o
o o o o
1 1 1
THO. o in
is -o cs o-
i i i i
•o (s in o-
o o TH CM
TH co in rs
TH
in ro TH m
N CK TH tM
1H TH
O-O CK IN
•T o oo in
ro
N in n is
ro •*• m m
-0 rt 00 -0
o TH TH rj
o o o o
i i i i
I- 00 N -0
o orj T
o o o o
1 1 1 1
iii o in o
TH TH n
o
O O O -H
o o o o
o o o o
N o TH n
n n n n
o o o o
1 1 1 1
o oo rs in
»• •« -o in
i i i i
n M «r *r
o o o o
T in •<> IN
TH TH TH TH
o- to n n
n •»• «r o -o >o
M3 -o rs rs
o o o o
o o o o
>o cs r-i n
ro ro «r «r
o o o o
i i i i
in o- v ro
TH co rs r>
o \ii in
o o o o
0 O O 0
•o n TH ».
-i m o -H
ri iii co rj
O 0 O -H
o o o o
1 1 1 1
•H n rs r-i
O O O -H
o o o o
1
in o in o
TH TH CJ
in
o o o o
o o o o
o o o o
o o o o
o o o o
o o o o
iii -o -o rs
o o o o
o o o o
1 1 1 1
-o rs co rs
TH o o n
o o o o
1 1 1
rs *r ro o
-o rs 03 o-
o o o o
o o o o
«r
TH TJ- CK O
1 1 1 TH
o in f «»
TH n T -o
TH r-i n in
i i i i
03 O O f 1
o TH ro to
o o o o
O O 0 -H
ro rs TH ro
ro ro
-------
O
o
oo
in
u. *
0
z
3
CC
ro
M
0.
O CM
SSOLVE
1 C
a
o
0Q
Z
O
tH
»- ^
:RCOLA
S B
GL 01
Z
tt
Ul
>
_J
U}0)
U)
a
a
in
u. t
0
z
3
cc
M
Z -I
M
z
a r-t
SSOLVf
«1 /
Q
0
REGION
AU t
O O O O
o o o o
o o o o
M M PI nos
o
o o oo
PI ro *o os
o o o o
o o oo
1 1 1 1
*• -o -o«r
*^ *4 vH ^4
O O OO
1 1 1 1
in o mo
*H ^H r-i
n
r*
TH «H TH O
OO O O
oo o o
pi PI ci rg
o o o o
o o o o
1 1 1 1
OsOs 0 -i
O O «H T*
ri M 0 00
O O O O
o o o o
PS os o n
o o ~* ^
o o o o
o n in -o
0000
Ps O » N
03 O -I M
O -« rt rt
PI n -o o-
o o o o
0000
1 1 1 1
00 O OS -0
0 -00
o o o o
1 1 1 1
in o in o
rt rt ri
n
-. -. ^. o
o o o o
o o o o
n PI PI PI
o o o o
o o o o
i i i i
0- 0- O **
O O ••< -I
CM n «r n
in in iii in
o o o o
in in r*) PI
o o o o
rf rf n o
in in ro ri
ro n ro fo
i i i i
in PS oo o
o o o -H
•H «r -o o-
03 Os Os O
n n n n
Os * 00 O
Ps 03 03 Os
» PS -i in
O O -H ^
1 1 1 1
* PI o- in
-i rf O O
i i i i
M «r -o o-
o o o o
o o o o
PS ON n «r
O O -H -1
o o o o
o- pi ro fl-
o o o o
Ps V PI -0
N 0- O 0 TH
-0 -0 » fl
ri PI- n PI
^ n -«in
•o ~o -on
» Ps rt -0
O O rf *H
i i i i
r^ 03 ^ rt
^ O O O
1 1 1
n in -oos
o o oo
o o o o
PS o- rf
o o o o
^ Os M M
in in x) -o
o o o o
PI V >0 O
O O O rf
o o oo
1 1 1 1
PS if) f -i n
o o oo
o o oo
1 1 1
in o no
r* ^ PI
in
r*
O O O O
o o o o
o o o o
o o o o
o o o o
0 0 0 O
1 1 1
* «r in n
n PS oo os
rt 1-1 ~1 TH
o o o o
n N PS y>
PI ri ri n
o o o o
1 1 1 1
IIT in o- o «r to n
o oo o
o oo o
1 1 1 1
in o in o
^ «M PI
Ps
*4
O O OO
o o oo
o o oo
n n PI ri
o o oo
o o o o
1 1 1 1
O -* -HCI
in rs co o-
«r o 03
i i i i
n to «r in
o o o o
O f 1 T -fl
03 OS OO
-< -^ PI PI
03 to ~00-
00 0- Os 0-
rj T in oo
o o o o
i i i i
Os 0- COO
o o oo
1 1 1 1
ri to * in
o o o o
o o o o
03 Os o ^
o o »»•*
o o oo
ro T in -c
o o oo
03 Ps V -0
». -i rom
O •* -4-1
^ n T >o
o o o o
o o oo
i t i t
CO Ps N -0
o o oo
o o oo
till
in o in o
-« -i M
03
227
-------
O
o
ca
in
u.
in nrs CD
00 0-0- 0-
T rs rt m
O OrH rt
1 1 1 1
n n *H o-
rt rtrf O
1 1 1 1
PI «r-o co
o oo o
o oo o
•0 COO- ^
o o o o
co •-» CM n
•< N ri CM
o oo o
O Tin N
•H HM «r
N «r fs o
O OO -i
o oo o
1 1 1 1
-o -on o
o oo o
o oo o
1 1 1 1
in oin o
o
CN
rf rt O O
O O O O
o o o o
o o o o
o o o o
1 1 1 1
0- Ct- O .H
00 03 O- 0-
o o o o
co oo o in
o o o o
N rj in rj
0- 0- N -0
0- 0- 0- O-
H
1 1 1 1
M -i oo »r
rf rt o O
1 1 1 1
n in -o o-
o o o o
o o o o
o o o o
oo o -< ri
rt n rj rj
o o o o
rf O- T 0
CO CO 0- O
O O O rf
N ^- N O
O 0 O -1
o o o o
fill
-o rx in CM
o o o o
o o o o
1 1 1 1
in o in o
*H
rj
oo o o
O 0 O O
o o o o
oo o o
OO 0 O
1 1 1 1
M PI ro «r
COO O -I
OO O 0
03 O -I -I
o o o o
1 1 1 1
030- rH o -o -o
o o o o
o o o o
CM is in o
CM ^T -0 03
-H «-t *H ft
•H n in N
o o o o
o o o o
1 1 1 1
* CM on
o o o o
o o o o
1 1
in o in o
5T
r-j
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 1
O- 0- 0- O
o o o o
^ ro to in
in -o -o -o
o o o o
*t o- n ri
in t in in
O O O 0
«• -«o in
o- 03 ro 0
(N
O O O O
O O O O
0 O O O
o o o o
O O 0 O
CK O TH H
O iH -< -^
0» CN OD 00
o o o o
f J O -O CD
rj rj tH ^
o o o o
o o- n
-------
o
00
o
oo
o
o
o
I
Q-
,
«T
3
i
TH tn •<• rs
s> » M TH
S3 S3 S3 S)
ro ro o oo
in o in o
-1 TH N
0-
IH
o o ro in
o s) rs ro
* o- o ro
in ro ro CM
00 H o OK
.H TH TH o
i i i i
CM O O 00
rs rs si in
o o o o
in TH o oo
^- v ro CM
n o in o
TH rH CM
0
*H
S3 -H in S3
00 TH » 00
rs si fs o>
ro TH ro oo
-o n «• ro
o o o o
1 1 1 1
CM * S3 o
O O O O
OK CM rs TH
-o -o n in
o o o o
in o in o
•4
M 00 0- »
S3 O CM *
TH to rs CM
s) in « n * I-
in N S3 CM
ro M cv CM
i i i i
S3 tH iH -0
03 S3 «r CM
rs rs rs rs
rs TH o co
ro ^ «r ro
mono
TH iH CM
TH
CM
o- rs rs «r
* -o o n
-o m tn ^
rs oo «r oo
S3 S3 S3 n
i i i i
CM 0- CM 03
S3 IO CM 0-
CM CM CM TH
O -0 0- 0-
rs rs rs is
in o m o
"1 r* CM
CM
iH
CM •* o n
ro oo -o o-
M rs «• -j
o- TO
in *• -o rs
i i t I
CM S3 00 TH
TH O 0- 0-
n o n o
S3
n o- * o-
oo ro ro ro
S3 o- m n 4r
*• O S3 S)
OK 0- 00 00
o o o o
till
CM o o in
03 rs S3 T
n ^ CM o
0-0-0-0-
o o o o
n o in o
TH TH CM
00
TH
O iH * O
S) O S3 CM
in o TH in
rs ro n o
tO O OK OK
IIII
TH n » S3
iiii
0- OK rO S3
«r ro ro CM
mono
0-
229
-------
o
co
<
o
a:
a.
co
o
o
o
_i
o.
CQ
in
u. «•
0
z
D
cc
ro
M
0.
o CM
SSOLVE
1 (
a
o
a
z
o
M
in
L. r<
M *0 -0 >0
0000
I i i i
»< -o n M
o ro co n
o o o •*
i i i
in o in o
•4 -1 (N
r4
o o o o
o o o o
o o o o
1 1 1 1
fx N N CO
o o o o
o o o o
rx r> N rv
rs in ^ CJ
o o o o
o o o o
co ••< ro »
o o o o
i i i i
ro n c* IIT
n ro ro <*•
o o o o
o o o o
in rv rv rv
o o o o
N O (N t
n ro ro n
-o in o- r<
rx o> o ri
*-l »^ Cl (N
ro «r in in
o o o o
1 1 1 1
r< in o- -<
rs co o- r-t
i i i -<
i
tO » 111 -0
O O O O
o o o o
^ n M to
o o o o
•< n » in
o o o o
o o o o
N CO (N CO
» ri ri ^
o o o o
0* to fs O
o o o o
1 1 1 1
o o rv n
•H ^1 O O
o o c o
mono
rt -H CM
ro
o o o o
o o o o
o o o o
1 1 1 1
rx N is co
o o o o
o o o o
rx PN N fs
0- -O « O
n n n n
o o o o
n r- o- —
o o o o
i i i i
o o- *-t rj
~o to « rj
0000
1 1 1 1
rj rJ ro *
o o o o
CO ^ » 03
O ^ -H -1
O O 0* CO
UT ^O -O N
CO * O ^
n rv N -o
rj n to ro
in co o n
0 0 - -<
i i i i
111 CO O CO
o to rv o
^ v^ *H rj
i i i i
in -o N co
o o o o
0 O O O
-1 (M M T
o o o o
CN o *-» in
*H rJ ro ^r
o o o o
CM in fs «•
o o- o *
r< O --1
IO CO CM -0
O O O O
1 1 1 1
co » in o
o o ti -o
0000
i i i
in o mo
»« ^ CM
V
O O O O
O 0 0 O
o o o o
1 1 1 1
rs rv co co
o o o o
o o o o
rs rs rx N
ro ** e>* o-
,« rt o O
o o o o
o n o rs
o o o o
iH *H iH 1-4
1 1 1 1
rs rs rs r j
N -0 111 >0
o o o o
III)
-i CM rj to
o o o o
-0 CO O -i
O O -1 TH
^ rs ro co
^ o
1 *-4 *-t ^
i i i
m -o is o>
o o o o
o o o o
r-i f j ro «•
O 0 O 0
CM -0 O -0
O O »-» •-!
o o o o
v rt in ro
ro M -i M
o o o o
ro co ro co
o o o o
i i i i
O CO -i <•
tl ** n O
o o o o
1
in o in o
»H *•! W
n
o o o o
0 O 0 O
o o o o
till
«• iii in -o
0 O O O
o o o o
fs fs 05 CO
>o co iii ro
•c in iii in
o o o o
o- ri t rs
ro o o o
o ** ** *~*
i i i i
-i r j tii
CO -< -< CO
f ro CM n
r) ro «r in
rs n co -o
O -i •-< CM
i i t i
<• 111 PO CO
rJ co in to
*-< ^ rj to
i i i i
ro ro CM rf
o o o o
o o o o
O O O -I
o o o o
t
ro n -o ui
O O -< 10
o o o o
1 1 1
111 0 bl O
^ r-t rg
03
0000
o o o o
o o o o
rj rj ri ri
o o o o
o o o o
Ll 111 >0 S3
n -o co *H
0- O O O
o c o -<
i-« r-i t rs
rj «r I,T I/T
O O r^ o
N -0 n ri
Ch C> T -•<
-i ro in «
Cl ri r i f i
I'll
-o -• o n
o <•« r j to
N co ro rs
•1 n in on
T ri rs 03
t "• rs t-i
-H ri n n
ro r-~ ^ n
co m t bi
to bi r- o-
iii -n o- o
rf n -3 03
i i i i
o t *o n
0- is T -i
ci «r >o co
i i i i
n ri ci to
o o o o
O O 0 O
cj «• n --o
O O O 0
•r oo o ci
CO to Cl CJ
0 r\ Cl to
bl 03 CO bl
•H tO Is t
-< " n bi
in 03 PS IH
o o o o
1 1 1 1
r o rs a-
S3 co r. ro
o ^ ro S3
i i t r
bl O bl 0
•-< ••< r-j
o-
230
-------
CO
REGION DISSOLVED N IN RUNOFF DISSOLVED N IN PERCOLATION DISSOLVED P IN RUNOFF
AW AO Al A3 A3 A4 A5 BO Bl B2 B3 B4 B5 CO Cl C2 C3 C4 C5
o o o o
o o o o
o o o o
CM n ri n
o o o o
o o o o
•r m in -o
in so rx o-
o o o o
T ri o
O- 0- 0- rt
rx o- rt
o o o o
in o- T m
O O 0 O
ro o
i
o
0
c
«r
o-
-0
000
o o o
o o o
1 1 1
O O 0
o o o
o o o
co o- ri
o o o
1 1 1
rx o- »
ro ro ro
l l i
0 O O
o o o
«r in in
0- ri f
ro ro ro ro
o
t
in
CO
i
o tn
ro ro
0 C
1 1
ro o-
i i
moo
a- o rv r>
ttii
O O 0 0
-* ri c; r j
0 O 0 C
T ro "•< iji
ro ro i
-------
o
o
ca
in
ii. <*•
0
z
a.
fO
0.
o CM
SSOLVE
1 C
M O
Ci
o
in
a
z
o
IRCOLAl
5 B'
(L a
z
M
z ri
a
tu
_j
O T-)
Ul «
01
o
01
in
o
z
u:
n
z
a (N
Ul cj cj cj
o o o o
n rs n •si
o 0"
0 0 0 O
iiii
s] T-. ro -o
o o .H cj
0 0 O 0
t I 1
in o in o
0-
0 O O O
o o o o
o o o o
1 i ! 1
o o o o
0 O 0 0
O- 0 CO Ps
ro ro ro ro
o o o o
o o o o
IIII
"0 rs rs in
Ell)
o o o o
cj ro ro ro
O 0 O O
•o ro o ~*Q
^o -^ iii
o o o o
1 ! 1 1
111 ^0 CO 0*
IIII
ro ^r ui in
o o o o
o o o o
^ cj co ro
O »H T-> CJ
o o o o
o in ^ rs
cj cj ro ro
o o o o
CD CJ CN CJ
O O O O
0 0 C 0
IIII
ui c«j in in
0 O O »f
o o o o
i i
in o in o
z
0
o
o
1
o
o
in
R
o
o
1
in
o
0
o
0 0
-r _< r J
III O SO
1 1
U) -> ro
0 T IT
i ' i
C C *-"
o ro in
-O -0 -0
o -« Is-
o n IN
T ~* T
I j »-.
»H r j ro
O O O
o o o
a- t co
i-« rj fi
000
^O r . ft)
0 O 0
fO T Ul
COO
ooo
r>oO
1 1 1
C Ul O
232
-------
o
00
cf
M
o
o
o
_j
o.
CQ
Ill
LL «T
O
Z
D
ct
ro
M
LL.
O CN
kJ U
>
_l
O
en
en rn
o
o
0)
z
o
:RCOLAI
S B*
z
u
CO CD
en
tH
Q
CQ
in
o
z
cc
ro
z
QM
SSOLVE
1 t
a
o
REGION
AU (
o o o o
o o o o
o o o o
I i 1 i
o o o o
o o o o
1 1 1
rs rx oo oo
Iii iii n iii
o o o o
o o o o
CO O O CX
ro «r >r ro
o o o o
1 1 1 1
cx fl- o in
•H CM CM -"
o o o o
rH IN N «T
O O O O
•o co rj ex
O O rH TH
O 03 CN in
CX rH 111 O
rH rH n
in CO rH O
cx o -o ex
rH r J *f -O
o o o o
1 1 1 1
rx IN co ix
rx roo o>
1 TH (N (N
1 1 1
•O 0- O TH
O O rH rH
O O O O
rH CX IX 00
IN IN ro ro
o o o o
1 1 1 1
oo r
O rH rH rH
O O O O
1 1 1 1
CN CN 10 *T
o o o o
0 O O 0
•o ri cx 111
rH r-i r» ri
o o o o
1 1 1 1
ri ro ix ro
rH rH n rO
O O O 0
ro in rx rn
rr ro
rH O O O
1
to in ix n
O O O rH
n CX O rH
rH rH rO 111
«r ix ri ro
O- rO O 111
rH rH ri
(N -0 -0 tO
o ro ri ro
to ro n o
o o -i ro
till
ix ro o ro
cx o- f ro
1 rH rO 111
i i i
ro ro ro o
o o o o
o o o o
ro
rH rH ri rO
•o o- m «r
•0 CO rH o
in rj LI rx
T 111 111 111
O O O O
-H ro o 03 ^ ex
TH ri iii ro
in ro -o ro
n o- -o n
TH TH ri ro
in cx rx co
CO rH |X V
03 CX rH O
O -H t IX
1 1 1 1
rH O lil IX
«• TH o ro
TH ro in -o
i i i i
rn TH ri ro
o o o o
o c o o
rs ro in rx
ro
in -o co o
O O rH -H
0 O O 0
i i i <
lil in ri ro
«r o in f>
o -< ri «
i i i i
111 O 111 O
cx
233
-------
in
L. -<
a
z
o
M
RCOLAl
B'
0_ »
03
a
I
0
z
ro
IH
z
QCM
:>
_J
O
03
03 -<
a
0
REGION
AU (
0.0 o o
0 O O O
o o o o
o o o o
o o o o
IIII
t in in -o
in in in in
o o o o
n ro * *r
o o o o
rx o ro CM
m o- in-o-
O O -1 *H
M n PI CM
iiii
ro so oo ro
O O O -1
o- ro o- n
o TH ^ ro
in ro N o
o* ro N ro
-i rt CM
in CM in «r
•o ro o- o
ro rx * in
o o •-< CM
iiii
03 CM 00 •»
rs «r ro -o
i i* CM ro
i i i
o o o o
0000
in CM o -a
CM ro ro CM
o o o o
IIII
03 --I * *«
>o o roco
0 -< rH -H
O M 0- 00
o -nn -i
ro n in -o
V S3 0* V
o o o -<
0000
IIII
rx ro rx o-
n in o- fx
000*4
iiii
in o in o
o
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 H
r
o o o o
10 -0 N 03
OO O O
-o co o ro
*-» 03 03 C>
rx o ro -o
»j CM ro in
o o o o
iiii
rx -I CM in
1 1 -H rt
i i
o o o o
o o o o
in rx rf T
•H *-t CM CM
o o o o
IIII
o- CM o in
V S3 00 O
O 0 O -1
O CM -0 CM
in o -o ro
CM ro ro t
^ S) GO O
0 0 0 ->
0000
IIII
oo vi 03 ro
o ~> PI in
0000
IIII
in o in o
•1 rt CM
o o o o
o o o o
o o o o
iiii
o o o o
o o o o
IIII
^ rx
o o o o
o o o o
IIII
-i » OD «r
o o o -i
o o o o
IIII
in o in o
M
iH
O O O O
0 O O O
o o o o
1
o o o o
o o o o
IIII
ri n ro ro
•o *o rs rs
o o o o
CM CM n ri
o o o o
IIII
in o o* N
rx 03 rs is
00 03 00 00
iiii
o o o o
CM ro ro ro
o o o c
•t a> >o
Ch --< ro in
n «• ui -o
o o o o
o o o o
IIII
^ o -H ro
o o o o
o o o o
1 1
m o in o
TH T-< n
in
O 0 0 0
o o o o
o o o o
IIII
o o o o
o o o o
in -o ^o -o
n n to ro
o o o o
!C K £ aJ
O O 0 O
IIII
o~ rs &. T
rJ »H ,H rj
ro ro ro ro
i i t i
o o o o
*t t in -o
o o o o
rj rx rj >o
ro ro 0 o
o o o o
0 0 O O
O O O 0
IIII
o o o o
O 0 O O
till
in in *o 0 0 O
O ^ •-» iH
IIII
o^ o tn «-<
t-< 0- 0- O
«r ro M ^~
1111
o o o o
o o o o
rj rj ro ro
o^ ^r in ^o
o *o
a- oo i> o
o o- a- o-
iiii
O O O -"
o o o o
rj TJ ro ro
o o o o
-o o* TJ in
n rJ m ro
o *-» o CD
n ro T
-------
o
o
"JO
CO
<£.
in
u. B
U)
M
a
a
in
u. i i
in co « n
•H -H (N [M
O 0 O O
rs rs ro rs
r> ^ ro «r
-i IN tM (M
to in *o is
o o o o
o o o o
1 1 1 1
r-i rt n- rs
o o o o
o o o o
1 1 1
in o in o
— ** CM
0-
*H
O 0 O O
O 0 0 O
O O O 0
1 1 1 1
o o o o
o o o o
1 1 1 1
1*1 ro * «r
n n to ro
ro IO rO rO
O O O O
-0 T -0 IS
O O O O
1 1 1 1
-o rs rs in
ts rs X3 >0
i i i i
o o o o
ro n o ~i n
O »-l »H yH
o o o o
i i i i
O- rO CO *T
•H rj N ro
o o o o
o- n <>• «r
rs o r
o o o o
o o o o
1 1 1 1
rn in (x in
n IN ri n
O O 0 O
^ o in r J
IPT rs co o
TH »H TH CJ
n in •« rs
o o o o
o o o o
1 1 1 1
•« is
N n n CM
o o o o
M o
n ri rj n
iiii
O 0 O O
o o o o
CM n n ro
O O O 0
in •« rs co
»H i-l TH *-l
in -H in CD
^ (N eg n
O O O ft
o o o o
IIII
O M -0 O
rH rJ ,-( M
IIII
rf rj rj ro
o o o o
o o o o
O O M -1
o o o o
O O O 0
IIII
1-4 iH *H *H
O O o O
o o o o
o ro f -o
o
o o o o
IIII
«r rs o ro
rf •* ri (M
o o o o
co ri in co
o ro tn is
ri ra r-i n
ro 0 ill 0- »O
o o- CO rs -O
r i ro T iii
iiii
o o — ri
o o o o
o c o o
i
-o c^ o rj
O 0 TH rf
o o o o
IIII
C- -i rj ri
^- >0 N 03
O O o O
N is ro C*
ro rs ,-•
-------
1
o
oo
— 1
( — 1
o
CO
•ZL
o:
o
o
0
i
Q_
j
1
<
u_
1
H— 1
1 — 1
fl -o
ro o n -o
CM ro ro ro
in o in o
rH rH CM
•0
CM
co «r o o-
rH N fl *!• ^
ro n ro ro
mono
rH rH CM
rs
-o -o CM ro
rs 03 in m
n rs CM 03
00 0- -0 -0
n n n n
i i i i
ro CM rs rs
in CN o- -o
co co rs rs
CM CO O O
-o -o is rs
mono
rH rH CN
CO
rs rH CD CO
CO O T O
m n «r r
o co rs -fl
rH O O O
o- o- ro -o
-------
o
GO
O
oo
Qi
0-
GO
o
o
o
i
CL.
CO
I
CQ
_j
en n
01
M
Q
B)
in
u. «r
o
z
ro
n
z
a CM
SSOLVE
1 t
w «
a
o
z
0 3
O
U
a
o o o o
o o o o
o o o o
i i i i
rH TH 1-H CM
in *• in -i
IS rH O CM
4T SQ CD *H
rH
CM ro If) 00
o o o o
1 1 1 1
rs CM ^ SQ
in o so *o
rH CM CM ro
1 1 1 I
*• in so rs
o o o o
o o o o
•O -O -0 CM
rH CM n «r
o o o o
f O- V i
O O -IN
oo o N in
rs CM TH *•
CM T -OO-
0* O TH O
o o o o
lilt
«f CO CO ^
IO CM TH to
O rH CM rO
1 1 1 1
in o in o
T-t
0 O 0 0
o o o o
o o o o
1 1 1 1
in in n in
o o o o
o o o o
CO CO CO CO
ro CM TH- -i
o o o o
00 00 N N
CM CN CM CN
o o o o
1 1 1 1
in -a co in
o o o o
o o o o
CM ro o N
o o o o
1 1 1 1
fM * CM M
M -0 O »
rt rH CM CN
1 1 1 1
in >o rs co
o o o o
o o o o
« m «• ^
CM M «r in
o o o o
^ n •<> o~
o o o o
in -H •-< -<
o rs oo ^~
*4 *+ C4 *•
O CM CM n
o o o o
1 1 1 1
M rv ro CM
i-t >0 CM C0
O O rH -1
1 1 1 1
mono
CM
o o o o
o o o o
o o o o
1 1 1 1
in in m in
o o o o
o o o o
N IX N N
M CM rH O
O O O O
rt n T -0
o o o o
o in o- -o
in so rs N
in -o rs oo
oo n rs -o
v o in o
n n n 0 -o >0 sQ
o o o o
o o o o
N rs is rs
so in >r n
o o o o
co o- -i ro
IS N CO CD
0 O O O
till
OQ ro rs o
0- Os 00 Os
o o o o
1 1 1 1
CM CM ro «r
o o o o
03 -H IO »0
O -H *H rf
•c n !>• in
T in in ^
T o n r-i
o t rs o
n CM n ro
n co o CM
O O ^ *H
i i i i
N in CM 1-1
o ro ^ o-
i i i i
•t in so rs
o o o o
O O 0 0
CM CM iH Os
»-i CM ro ro
o o o o
O O O 0
o so ro rs
so in so co
o o o o
-! in o T
0 0 O O
1 1 1 1
in in 1-1 in
i« o ^< ro
o o o o
1 1
in o IIT o
in
o o o o
o o o o
o o o o
i i i i
>T in in in
0 O 0 O
o o o o
03 03 03 CO
111 -H 0- CO
in in f *•
o o o o
rs OD co oo
o o o o
1 1 1 1
in OD in T
o «r in so
o o o o
1
in
o o ^ ro
i I I I
mono
rs.
0 0 O O
o o o o
o o o o
iiii
in so so so
o o o o
o c> o o
sO -0 sQ sO
n -< o o-
N u n ~4
O C) O O
rH n rs
OD rs in EN
rs co o- o
rH rH rH OJ
IIII
rs ro so os
o r-t ri ro
o in o- in
CM ro so o
rH sQ in SO
in n Os «r
rH n rt ro
t in o- o
O CM (N 0*
-------
o
o
oo
CO
1 O OOO
in i o o oo
1 O OOO
i
u. i o ooo
0 1
en i rf ino«r
at «-< i o o^^
a i o ooo
i i
1 00 111 WOO
O 1 CM -O O*
1 1 1 1 1
1
i in OOCMCN
a i o O-ICN
i
Z 1
O 1
M I o tvtvco
_i i
O 1
U 1
a. a i
Z 1
M 1
A 1 -0 00 O-O
LU 1
^ 1
in 01 i o rtM»
U) 1 1 1 1 I
a i
i CM ino-o
m i o o CN'O
i i i i i
i
i
i
in i o o oo
1 O O OO
i
i
i n * to *
U. * 1 *-i CN tO *
0 1 0 000
Z 1
a: i N oo oin
i
Z 1
i n
0- 0- 0- 0-
N M 0- -0
nrvo-cN
O O O i
1 1 1 1
1 1 1 1
oooo
oooo
T-I t-l CM M
OOOO
O O «H «H
^ •« -o
o to to <»•
O 0 0 -H
1 1 1 1
in o in o
«H
OOOO
oooo
oooo
1 1 1 1
oooo
oooo
in in n -o
o o o- oo
mint v
oooo
n n -o tv
oooo
1 1 1 1
in to » to
0- O O O
i i i i
oooo
•O 00 O CM
in * o
IO -0 00 0-
oooo
1 1 1 1
-0 00 O CM
1 1 -1 rt
1 1
oooo
oooo
r* rv to oo
O O *H -1
oooo
oooo
« n o- t
t n to in
-O 0- CM »
O O rt -H
OOOO
1 1 1 1
O O CM *
OOOO
1 1 1
in o in o
CM
oooo
oooo
oooo
oooo
oooo
O-O-O-O-
o o o o
CN ** *4 *-)
CN CM CM 04
oooo
1 1 1 1
o n n *
O-O-O-O-
1 1 1 1
•H CM CM CM
oooo
oooo
n m CM <-c
n...
M in rs o
«*«••*
oooo
1 1 1 1
1 1 1
oooo
oooo
oo n N CM
O rt rt CM
oooo
is n »r T
CM n * in
oooo
in in CN -o
n n v n
oooo
-0 00 O M
O O -I rt
oooo
1 1 1 1
O 0- F> -H
O O CM »
OOOO
1 1 1
in o in o
n
oooo
oooo
oooo
1
oooo
oooo
CN CN 01 *H
oooo
N -0 -0 -0
CN n CM CM
oooo
1 1 1 1
CM <• 71 -Q
O O O 0-
00 CO 03 N
1 1 1 1
oooo
oooo
r-, o- o- o-
n -o rs oo
o- o rs «r
CN -0 00
-------
oo
i
CQ
in
a. »
o
z
D
or
M
M
0.
a IN
SSOLVE
:i c
a
0
in
n
z
0
h-t
H
rs o CM in
i i i
CM M ^- in
o o o o
0000
in -H -o -i
O rH rt M
o o o o
O O 0 O
00 If) 03 M
in tn in is
o o o o
-0 0- (M »
O O T-( ~»
O O O O
1 1 1 1
» M rx N
O O *-l M
o o o o
1 1 1
mono
o
(N
O O O O
o o o o
0000
1 1 1
0000
o o o o
M * ^ in
00 00 CO rs
M M M M
o o o o
o o o o
M M M M
o o o o
1 1 1 1
N rx -o •«•
in in in in
i i i i
(N CM M M
O O O O
«t- in -o rs
o o o o
« 0- IS V
tn in •<) rs
f -o is rs
n M in is
M in -o rs
o o o o
1 1 1 1
in M OD -o
-0 00 0- •*
1
CM M <• in
o o o o
o o o o
in o in o
O •* ~* CM
O O O O
0 O O O
« -o rv o
in «• T in
o o o o
•0 03 rt M
O 0 1 TH
O O O O
1 1 1 1
n * in o-
O O -1 CM
o o o o
1 1 1
in o in o
CN
o o o o
o o o o
0 O O 0
1 1 1 1
o o o o
o o o o
in iii -o -o
M (N CM IH
M M M M
O O O O
N N rx oo
«!•«••»••«•
o o o o
1 1 1 1
M .H CN CM
in in in in
i i i i
O o o o
+ v nn
o o o o
o M in rx.
M M M M
Ox in -0 CD
M in -o ix.
CM M ^ in
o o o o
1 1 1 1
«r -o -orx
rx. 03 o- o
CM M M «T
o o o o
o o o o
0- * o in
O rf CM CM
O O O O
o o o o
oo rs 03 in
in m in -o
o o o o
rs o (M v
O rt -I rt
o o o o
1 1 1 1
-0 O M IS
O O rt N
O O O O
1 1
in o in o
CM
CM
O O O O
o o o o
oo o o
i i i i
o o o o
0 O 0 0
in in in in
oo rx. rs rs
*H *H 1-4 r-l
O O O O
CM t -o rs
in in in in
o o o o
1 1 1 1
*-< TH *H TH
M M M M
1 1 1 1
o o o o
M V * *•
O O O O
•0 00 O CM
CM CM M M
in co ox co
CM M T in
CM M M M
O<5 O O
1 1 1 1
rs -H ^ in
-0 00 0- O
1
CM CM M *•
O O O O
o o o o
in o- «• oo
o o -i -<
o o o o
O O 0 0
rs in CM ^>
CM CM CM CM
O O O O
in 03 -H M
O O -i ^
o o o o
1 1 1 1
rs -o CM M
o o o o
o o o o
1
in o in o
M
CM
o o o o
o o o o
o o o o
1 1 1 1
o o o o
o o o o
•0 -O -0 -0
in » r
^ o -o -c
o o o o
-o o- n in
O O -H ^
o o o o
1 1 1 1
^ IS -o O-
o o rt n
o o o o
i i i i
in o in o
•0
CM
0 O O O
o o o o
o o o o
o o o o
o o o o
M «r «r tn
in
-------
O
00
o
oo
o
C_5
o
_i
D.
CTl
I
UJ
i
CO
•=c
in
U. V
0
z
ct
M
o.
Q CM
D I
smoss
Q
o
CQ
O
:RCOLAI
I B^
CL m
z
a
a
en a
en
a
a
in
u. *
o
•z.
D
a.
n
z
Q CM
UI 1
>
_l
O
cn
U) TH
Q
0
REGION
AU t
O O O O
o o o o
o o o o
1 1 1 1
o o o o
o o o o
1 1 1 1
rs rs oo oo
in in in -o
o o o o
0000
CM CM CM CM
o o oo
1 1 1 1
in in oo n
O O O TH
o o o o
i i i i
CM CM n -0
o o o o
rs TH rs o
0 TH THCM
o ro rs oo
o ro rs «r
TH TH TH CM
CO 0- (N -0
CM -0 -0 -0
«r in rs o
TH
CN to in 0-
0 0 OO
1 1 1 1
o o CM ro
o- -o in -c
i TH CM ro
i i i
r
o o o o
o o o o
o CM ro T
o o o o
1 1 1 1
in NO oo o
O O O TH
o o o o
o ro *o °o
ro ro «r «r
ro ^° ~o rs
o o o o
o o o o
i 1 1 1
CM CM TH CM
O O O O
o o o o
1
in o in o
TH TH IN
ro
0 0 O 0
o o o o
0000
1 1 1 1
o o o o
0000
1 1 1 1
rs rs N rs
rs co co oo
o o o o
o o o o
TH CM «r in
•0 -0 -0 -0
o o o o
1 1 1 1
to
TH n ro »o
i i i l
IN IN TH TH
o o o o
o o o o
1
TH o in is
in rs co es
o o o o
i i i i
rs o CM in
O -H -H tO
is TH
TH rs o CN
O O IN «•
1 1 1 1
in o in o
TH TH (N
•o
o o o o
0000
o o o o
1 1 1 1
o o o o
o o o o
1 1 1 1
rs rs N rs
M ro » o
N in «• »
i i i i
to *o ON ^
0 O O TH
TH O IN fO
TH N to in
o -o in rs
N o n in
TH TH
TH o. in in
in o co m
CN ro to ^
O O TH to
l l l l
to o in rs
-o TH rs in
i TH TH rj
t i i
N CN TH TH
o o o o
0000
^ co ro TH
<• in rs co
o o o o
1 1 1 1
o -o o in
in a) 0 -0
in in in ~o
o o o o
TH IN «r -o
o o o o
1 1 1 1
T o in in
O 0* CO 00
n IN CN CN
1 1 1 1
TH CN CM tO
O O O O
rs o> TH t
0 0 rH TH
in TH -o IN
to O
TH TH TH CN
TH CM
-------
O
o
cr>
i
ca
i oooo
n i ooo o
0 ! oooo
1
U. V 1 OOO 0
O 1 OOO O
Z 1 1 1 1 1
1
i «rfsCMx
at i oo*< CM
i
Z 1
O 1
M | -irsrs CM
« a i »H -H CM n
-i i
0 1
U 1
a i rocoovro
a. n i 0*0- -o
( .H *H «H CM
Z 1
hH |
i o^m m
n i co in M *o
U 1
:> i
0 -« 1 • • • •
0) £0 1 O " i «r -o rs co
-i i
O 1
in i ^ «o CK 10
U) M 1 O O O -<
a i o o o o
i i i i i
! o- m n 10
O 1 CM •« ro •«
* ! 00 -IN
1 1 1 1 1
03 i in o in o
U 1
Ul 1 O
K 1 -i
oooo
oooo
oooo
i
oooo
oooo
i i i i
in in -o <
M CM M M
oooo
in CM CM ro
ro to ro ro
oooo
1 1 1 1
o* CN 0 00 O CM
O O -I 1H
O *0 CM O
* v in m
* o n *H
n n -o N
oooo
1 1 1 1
CM o in n
n
t in\n in
r*i o >o
CM CM CM CM
OOOO
•» n -o co
n n n in
oooo
1 1 1 1
-o »H «r rs
n n ro n
ro ro ro ro
i i i i
oooo
n *o *o CD
oooo
•o ro o- •«
ro ^ «r n
o> cj o rj
co ^ n •«•
-* ^H n ro
oooo
1 1 1 1
-o o- in n
CM ro n rs
i i i i
CM ro «r n
oooo
oooo
rs co o* o
o o o -i
oooo
1 1 1 1
in *H CM CM
OOOO
oo CM ro in
o in o- ro
ro ro ro «r
ro o *o *c
00-1-.
N CM CM CM
OOOO
•T rs 0- rH
N rs N 00
oooo
1 1 1 1
rs CM n rs
•« n n in
ro ro ro ro
I i I i
oooo
n ~o rs co
oooo
o- 0
00 CM -0 O
n ro ro T
CM
till
oooo
ro ro ^
-------
CT>
I
CO
in
u. «r
a
z
3
Lt
M
M
a.
a CM
>
_i
o
in
in TH
a
o
in
0)
z
0
M
0
O TH TH TH
o o o o
TH in M3 o*
CM in co TH
M n n 0 N N rs
o o o o
1 1 1 1
o rs co
o o o o
TH -0 0 «•
ro ro ^ *•
«r rs -a ro
N ro r in
ro «r «r ^~
ct. rs o CM
0- O TH TH
o o o o
1 1 1 1
rs *r TH o-
I I I I
TH CM ro 0
CM
O O O O
o o o o
o o o o
o o o o
o o o o
1 1 1 1
ro * «r in
T v ro ro
ro ro ro ro
o o o o
<• N TH ro
-o >o rs rs
o o o o
i i i i
co rs ro CM
i I i i
O O -H TH
o- * o o
o TH CM ro
CM rs TH TH
in in -o -o
rS CM 0* TH
O TH O O
ro N CM o
O O -H CM
1 1 1 1
CM CM CM O
1 1 1 1
TH .H O CM
O O O O
o o o o
1 1
o- ro o 03
0 O 0 O
1 1 1 1
rs v TH co
-0 CO O TH
O O TH TH
oo TH -o «r
OK NO CM 00
TH n r*} ro
ro in co CM
0 O O TH
o o o o
1 1 1 1
in o N *o
O -H -H CM
o o o o
1 1 1 1
in o in o
TH TH CM
rs
CM
242
-------
O
oo
O
O
O
_l
D-
*
o ^ rs in
TH o o* CK
- -io o
i i i i
CD rs TH rs
o- rs m -H
in in in in
CN CM CN CM
ro ro o CD
in on o
-I** CM
o
TH
-H o ro CM
CN ••* rs o
ro o* o ro
CD «T TH O
ro -HO oo
I I I i
-H -iro is
in 01 o CM
M M (N CM
0* £N ts *H
-o -on in
o o o o
in oin o
-
CO
TH »H O O
1 1 1 1
co n O
•o ro rs rs
«T CN T CO
>o «r TH CM
CM o o- 00
I t i i
o* >o in ro
(N O 111 O
co co rs rs
TH in M o
n (N CN (N
in o in o
TH O rs -O
o co in M
oo is o in
in n n in
O 0- 00 IS
r-i O O O
i i i i
in o T o-
o o co *o
>o *o in in
0* (N *0 0*
O 0 0- CO
•H « O O
in o in o
(N
00 00 M) -0
M CN 00 fs
nin o- in
in n -o 01
M n n r-i
i i i i
rs M in IH
fl- r^ rs ^i
•o ~o in in
n (N n 0 IS M fS(
co -o — rs
O. -H •« -I
rs co T oo
•o -o •« in
i i i i
-0 "i -H CO
(N 0- -0 rt
fs -0 -0 -0
O -0 0- 0-
rs rs rs fs
mono
N
r*
O 0* O ^
00 M M OD
IS N 0- -0
o- fi v n
«r in m in
i i i i
oo n * o-
rj o- rs *r
o »• o> tx
o «• ^ M
in in in in
in o in o
ro
n rs o- o
UT O rs CM
03 M CK Pv
o -* rj CM
n PI M ro
i i I i
n co n CN
oo n o
CM ro oo ro
o- oo in in
-4 o 0- 00
i i i i
•0 -0 0* -0
ro rj o* •«
CK {> CO CO
to CM O 03
CM N CM ^
mono
to
to N ro o-
O N CM tO
•o •« o in
rs to t v
i i I I
n in rs rs
ro CK >o ro
M (N CM M
o to o -o
~o ~o -o n
mono
^
o -o n
•o oo ro o>
CM 0- 0- 00
i i i i
to ro CM ro
(N O N t
*o >o in in
» IS -0 -0
in n n n
n o n o
n
•o rs o- *
•oo -o «r
n ro ^ *•
ro ro ro ro
i i i i
O T O -0
rj o- rs 0 O CM CM
co o ~a
rj -o co o
— 00-0-
n o in o
•0
•o -H ro o-
00 ^ -0 CM
O CO O C*
n ^-n ^
i i i i
CM in CN rs
n o is to
CM CM *H *-(
IN M CM TJ
r-j ro n in
in n n in
in o n o
i ~ «
n
CN
n oo rs n
in CM o ro
T IS CM 00
0- CM IS CD
o o o- o*
i i i i
n o- •« •«
ro o co n
rt rf 0 0
CN o 03 rs
-H ^ O O
mono
•0
TH
rs 03 o- 05
*c -o -H n
* **m -a
ro *H n >o
n » n o
i i i i
o -o n n
rj rs ro o
fs •<> -0 'O
o «r -H rs
>o *o *o in
n o n o
rs
" ro n »
•o n •« oo
ro in -H rs
-H n o CM
n n ro ro
i i i i
o- rn co n
O -0 O N
v n ro CN
to o n >o
CM ro ro ro
mono
-H -1 CM
-0
CN
o- oo «r o>
CO O- CM ^
•o »• n *«
CN o oo rs
ro ro CN n
i i I i
ro o- n n
CN CK rs *•
O 0- 0- 0-
» •« v »
ro ro ro ro
n o n o
N
TH
03 ro o o-
03 O IS 00
rs o » o
00 0- -0 O
n n n n
i i i i
o rs rs
n o n o
CO
TH ro co n
co ro VH ^
to n o >o
>o TH ro "O
in MJ -o ~o
i i i i
TH r , rs *o
n co o ro
03 N N *0
co -o
-------
o
o
00
CJ3
o:
D-
co
a:
o
03
in
u. o -o
o o o o
in in T o in o- rH
-o CM in -i
CO rH
CM in &• CK
-o rs oo c-
O £D in CD
CD t rt ^O
n n T O 0
o o o o
»H » M 0-
^ »0 CM »H
-H ^ rg M
O * O -0
rH TH n CM
o o o o
i i i i
«r n o> •<>
o T o- rs
o o o -i
i i i i
m o if) o
-1 rt CM
»
O O 0 O
o o o o
o o o o
1 1 1 1
o o o o
o o o o
rs rs rs rs
o- co rs rs
o o o o
o ^ ro T
-0 >0 -0 -0
o o o o
i i i i
n CM n CM
^ ,H rH *-4
i i i i
CM *r in rs
o o o o
•H -0 f) O
CM rH 0- >0
in O 0- CO
if) -o in in
o o o o
1 1 1 1
CM O O O
O O O O
1 1
in o o- rs
o ^ ^ ro
o o in is
rH
CO -H -0 O
n o- N co
rH rH CM rO
to o- in o»
O -0 rH 111
in rs rn in
rH rH
ro co CM o-
1 1 1 1
rs CK T ro
rs -o CM ^~
1 1 1 rH
0 0 O O
o o o o
0- O CO W
O O O rH
O 0 0 O
* rH Os CM
oo rs oo f
O rH CM *
rs *> rs ^i
if) n rH co
in rs rH -o
rH rH
&• rn CM ro
O rH rH rH
0000
1 1 1 1
O -0 IN O
o «r in oo
o CM in o-
i i i i
in o in o
rH rH CM
-0
o o o o
o o o o
o o o o
i i i i
o o o o
o o o o
rs rs co co
o- rs in T
o o o o
o- rH CM ro
in -o -o -o
o o o o
1 1 1 1
v ro n n
i i i i
in o* in o-
O O rH CM
>o ro «r o-
N -0 00 rH
oo rH »r oo
oo -o •
i i i i
rs ^ in f
in ro CM CM
l i l i
O 0 O O
o o o o
rs f n -o
O O rH CM
O 0 O O
rs oo -o «r
•0 rH CD -V
O rH rH CM
CM o -o in
a) ro rs o
ro T in oo
o* n "O n
O rH rH CM
0000
1 1 1 1
in oo oo rH O
-------
O
o
ca
REGION DISSOLVED N IN RUNOFF DISSOLVED N IN PERCOLATION DISSOLVED P IN RUNOFF
AU AO Al A2 A3 A4 AS BO Bl B2 B3 B4 B5 CO Cl C2 C3 C4 C5
O O O O
O O 0 O
0 O O O
CN CN CN CN
O O O O
O O O O
*o -o rs is
-o rx oo o
o o o o
CN rs o in
rH CN CD CN
CN CN ro f
rs CN CD CO
O rH T-t TJ
i i i i
in ro rH CN
i i i i
O O O 0
o o o o
CN in ro CN
o ^ CN ro
o o o o
O rH rH rH
CD O M rH
O O rH CN
CO rH in rH
O rH rH IN
O O O O
1 1 1 1
oo -o o o o
i i i
o o o o
o o o o
Ps CN N CN
O rH rH CN
O O O O
0 O 0 O
rn ro *o o
^ HT in oo
o o o o
in CD rH <«•
O O rH rH
O O O O
1 1 1 1
0 CO CN rs
O O 0 O
•o CN «r CD
ro ro in rs
o o o o
in CD CN >o
O 0 -H rH
0 0 O O
1 1 I 1
ro 03 rH oo
O O O 0
1 1 1 1
in o in o
rH rH CN
0
o
o
1
ro
0
0
O N N
in v T ro
o o o o
o o o o
1 1 1 1
co <>• CN in
in o
0 CN O-CN
O O O O
o o o o
1 1 1 1
CD 00 CN in
in in ^o -o
CD CD 00 CD
1 1 1 1
rH CN rN n
O O O 0
-------
O
o
co
<
n
u. T
o
z
"n
M
a.
QM
ISSOLVf
:i (
a
8
a
z
o
:RCOLAI
t B'
0.01
z
M
z«
s
(00)
(A
t-»
a
O)
n
u. j
I
M
Z
a CM
SSOLVE
>1 *
a
o
REGION
AM (
OO O O
OO 0 0
oo o o
1
NCM N M
OO O O
oo o o
nn -o -o
oo o o
nn n n
OO O 0
09 CO 09 0»
MM MM
OO 0 O
1 1 1 1
rsM -o oo
n< * »
rsrs rs N
i i i i
MM n n
oo o o
nrs rs 09
oo o o
0-0- 0- OK
n-o rs co
co •« n n
no- CM n
nrs co o-
oo o o
i i i i
09 00 0* CD
on -o o«
1 1 1 1
Mn » -o
00 0 O
oo o o
S28R
oo o o
N-O -0 •«
Mn * n
OO 0 0
OM N «•
MM n n
oo o o
-00- (N -0
OO -I -•
OO O O
1 1 1 1
TN -0 «H
OM » rs
oo o o
1 1 1 1
no n o
~ -H M
Cf
V4
O OO O
o oo o
0 OO O
1 1
N MN M
O OO O
O OO O
n n-o -o
09 oo oo rs
n nn n
0 OO 0
N ON rs
M MM N
O OO O
1 1 1 1
N 0- CO 00
0-0-0-0-
n nn n
i i i i
M n » n
o oo o
•fl COO ~
0 0~* •*
*• o-n o-
•0 N 0- O
«•«
ONTO
•O O» 0-
T NO- -H
ooo •*
1 1 1 1
0. »0 rH
o- nN -
1 1 1
M n* -o
o oo o
0 OO 0
n o-o n
O **•* N
o o o o
n n n o
n Tn N
0000
»H -N 09
fs N 00 O
O OO -i
n o>M n
O O T< TH
ooo o
1 1 1 1
§T-O n
*n •«
o oo o
1 1 1 1
n on o
•4 >4 N
o
M
§s§§
0 O O O
1
N n nn
o o o o
o o o o
T n in >o
oo oo N N
n n n n
0000
T T » T
N N NN
0 O 0 0
1 1 1 1
0- M 03 in
«r n OK
N OK 0900
M -0 0- M
T oo M n
0 O - -
i i i i
•< T -0 N
03 0 M in
i i i
N n «• -o
o o o o
o o o o
n o •« N
O rt ^ (N
o o o o
•o n •« o-
n T n •«
o o o o
N CD T O
•« in -o co
o o o o
809 ~in
O ri rt
o o oo
1 1 1 1
T< -1 030-
O ^ CM*
o o oo
1 1 1 1
mono
IH ^N
*H
CM
OO 0 O
oo o o
o o o o
1 1 1 1
nn n n
0000
0000
-0-0-0-0
oo o o
no> CM n
nn T T
nrx n oo
•fl 03 O n
*n N oo
oo o o
i i i i
04 o n
o- o CM n
i i i
CMn n n n r^
n n n n
n n n n
i i i i
«-4 ft lH T4
O O O O
n -o N N
0 OO 0
N -003 -I
n n n *
» *• o. CM
n is oo o
* nn n
0 00 O
1 1 1 1
•0 IS *• 0-
09 0 CM n
1 1 1
CM CM n »
o o o o
0000
m o- n
0 OO ^
o o o o
i i i
n on o
«•« »H CM
n
CM
o o o o
o o o o
o o o o
1 1 1 1
n n n n
o o o o
0000
-o -o -o r-
oo rs *o -o
CM CM CM CM
O O O O
M » n -o
n n n n
o o o o
i i i i
03 o- o CM »
n N oo o
o o o •<
i i i i
O 09 CM CM
o CM nrs
1 1 1 1
CM n * n
o o o o
0 O OO
•fl CM O -O
O <1 CM CM
O O O O
CM -0 CM 09
n rt CM CM
O O O O
*« » •« n
09 09 09 O
O O O -i
N n n 00
0 rt -I *
O O O O
1 1 1 1
n rs *• n t
i i i
Mn » in
o o o o
OO 0 0
NM oo n
O *•< •-• CM
o o o o
»O N -0
CMn n *•
o o o o
O- IS 09 03
» » « n
oo o o
r-o T N
O
o - M
•« -o •« rs
o o o o
1 1 1 1
O 0- 030-
o rs » n
rs -o -a -o
i i i i
•o o •an
O •< *« M
M -H CM -«
•H M n »
00 M CMn
n -o -on
^ ^ n»
•H M M •<
co -o in -o
o -< Mn
i i i i
o> 09 no
•0 rs 09 09
^ o »* *
o o o o
o o o o
1 1
rt » «O
O O rH CM
o o o o
1
rs M N 09
oo n n r>
o «-< *^ —<
n •« »• »
» n -o co
—1 «4 *H *H
rs o n CM
O -H — (N
o o o o
1 1 1 1
•o o rs »
o CM n n
o oo o
i i i i
n o no
rf Tt M
N
CM
246
-------
O
OO
LJ_
la-
o
o
o
_1
0_
CM
r—
I
in
u. «r
0
z
D
oc
M
t-t
fl.
Ci CN
SSOLVE
1 C
Q
o
03
§
M
-to,
0
u
QC
L4- ID
z
M
0)
Q
UJ
:>
_i
tn to
en
M
Q
0)
in
u. «
a
1
to
VH
Z
a N
SSOLVE
»1 I
Q
O
REGION
AU <
o o o o
o o o o
o o o o
1 1 1 1
o o o o
o o o o
o o o o
1 1 1
rs 00 OD 00
in in in -a
o o o o
o o o o
in in o n
OO TH TH
OO O O
1 1 1 1
O IN 0- M
MTH «r M
o n CM O M
•H in rf o
O O»< CM
1 1 1 1
iii o n o
T* !-< (N
(M
o o o o
o o o o
o o o o
1 1 1 1
o o o o
o o o o
0000
1 1 1 1
N rv r* N
-0-0-0-0
o o o o
o o o o
0- O -< CM
o o o o
1 1 1 1
rf -0 111 M
«r M to to
o o o o
1 1 1 1
rt (N N N
o o o o
o- CM « r-
O ** T^ 1-»
00 M N O
-c r
-0-0-0-0
o o o o
1 1 1 1
N <• -H to
CM CM M CJ
»H *H *H i-<
1 1 1 1
CM to in -o
o o o o
O «T O O
•^ ** r-4 to
CM »H O 0-
«r in -o -o
in r< IO rf
^ » m -o
CM CM CM CM
to m o in
O O -o -o
0000
1 1 1 1
^ *H oo rs
-o -o in in
CM n n CM
i i i i
CM (O f ~0
o o o o
o- to rs to
o *H *-< rj
o co -o to
o o
TH rt n cj
CM «f 00 CM
o o o •*
1 1 1 1
o in n in
« in rs o-
i i i i
CM to ^ in
o o o o
o o o o
» rs oo o
~< ^ ^ n
0 0 O 0
1 1 1 1
-H 00 N 00
n M to «•
o o o o
tO 00 CM 0-
^ oo in -H
CM
o o o o
1 1 1
in o in o
-> « CJ
00
0 O O O
o o o o
o o o o
n n n CM
o o o o
o o o o
1 1 1 1
oo o- o n
to ro » *
o o oo
0- to CJ O
M n CJ CM
0000
1 1 1 1
V 10 0- »
0- ^ 0- -0
» in in •<>
1-1 *H TH T-t
1 1 1 1
o- o- in -H
o — ro »
» oo rs in
CM
-------
O
CJ
CNJ
I
QQ
in
ii_ ^r
o
z
3
a.
ro
M
CL
Q CM
SSOLVE
:i c
Q
O
in
a
z
o
IRCOLA1
S B'
CL 0)
Z
M
ID
LU
^
_j
u) a
in
HI
a
CQ
111
h_ «r
c^
ro
Z -i
Z
a CM
l\
rnoss
a
0
z
0 3
a
u
tL
o o o o
0000
o o o o
o o o o
o o o o
1 1 1 1
•o o rs rs
ro ro ro o ^
in in in •«
t f i i
i i i i
0- «• (M fM
O CM N «T
rf (N ro in
1 1 1 1
O O M N
O O O O
O O O 0
*T O O iH
CM ro ro CN
o o o o
1 1 1 1
IS *T lH iH
o >o *r in
-1 rt CM n
oo rs •»• o
co o- m ^r
IH
^ rs (N n
o o -< CM
o o o o
1 1 1 1
•0 0- ** rs
n co o o
o o r-i o o o o
«r in in in
o o o o
in N co ^
O O O -I
co fi rs (o
in rs co o
n o- CN -o
O O* rH
CM ro IIT -o
o o o o
1 1 1 1
-o in o- o-
«r ^3 co »H
i i i -«
CN n
o IH rj ho
o o o o
1 1 1 1
in o in o
n
o o o o
o o o o
O 0 O O
O 0 O O
o o o o
1 1 1 1
«r in in -c
CD CO CO CO
ro ro n ro
0 0 O 0
T* TH TH 1H
CM CN CM CM
o o o o
1 1 1 1
>0 CD >0 M
-0 -0 -O -0
O O 0 O
in N co o
o o o *-»
t -< CO CO
-0 CO 0- 1-1
1-1 CJ CO 111
in co o ro
CN ro in -o
o o o o
1 1 1 I
o in 0 O
*H n TJ rj
rj *r o oo
o o o o
O 0 O O
1 1 1 1
ro o -H -o
o " n ro
o o o o
1 1 1 1
in o in o
t
o o o o
o o o o
o o o o
0 O O O
o o o o
1 1 1 f
•r t in in
ro 10 ro ro
o o o o
CO CO O~ IH
ro ro ro o *o
CM ro o fs
ro n so o*
n ro t n in
ro ro ro ro
o o o o
rs o- ^ «r
O O -i rt
in CM o- -o
ro » *• in
1-1 -H rs o-
co o -i n
rf ro n rs
o o o o
1 1 1 1
o in ro >o
ro ^ -o co
i i i i
n ro T in
o o o o
o o o o
>o rs rs rs
0 O 0 0
o o o o
1 1 1 1
in o- «• o
o o o o
-< o o -i
in o t co
ro v v «r
CM * *0 CO
o o o o
o o o o
1 1 1 1
V CM CM O
O O O O
o o o o
1 1
in o in o
IS
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 1
in in 'O -o
in in in in
ro ro ro ro
0 0 O O
-0 -0 -0 -0
o o o o
1 1 1 1
co co CM in
in in -o *o
CO CO 00 CO
o o o o
T -o rs oo
o o o o
o- ~o ro ~
ro «• in -o
ro n o* v
-0 CO 0- rf
T-I n CM «r
o o o o
1 1 1 1
-i
-------
O
o
C\J
00
in
u.«r
o
z
D
ro
0.
QCM
SSOLYE
1 C
o
o
z
o
O O O O
o o o o
o o o o
1 1
0000
o o o o
1 1 1 1
in UT •o -o
rx rx rx rx
n CM CM n
o o o o
N -0 N IX
CM CM CM CM
O O O O
1 1 1 1
rx o- oo m
0-0-0-0-
n n n in
i i i i
CM ro «• n
o o o o
n co o ro
o o -i -i
oo CM -o ro
n ix co o
—i
-0 rf -i O
t rx o- CM
CM V -0 ix
0 0 0 O
1 1 1 1
0- 0 IX O
«f ix o- ro
i i i -i
i
CM ro ^- iii
o o o o
o o o o
•« rx fx -o
o o o o
o o o o
1 1 1 1
ro -i CM -o
o o o o
10 0- -" t
•o o- ro -o
CM CM ro ro
ro t -o o>
o o o o
o o o o
1 1 1 1
-I Nix IO
o o -i n
o o o o
1 1 1 1
n o n o
rf -i N
O
CM
O O O O
O O O O
o o o o
1
o o o o
o o o o
1 1 1 1
r
o o o o
1 1 1 1
ro -o o- ro
o o o o
0- CO •» O-
03 CM -0 0-
ro » T f
CM ro ^ in
oo o o
o o o o
1 1 1 1
<• CM CM N
O O O O
oo o o
1 1
mono
»H -t CM
CM
CM
O O O 0
O O 0 O
o o o o
1 1 1 1
o o o o
o o o o
•0-0-0-0
rx ix rx rx
o o o o
o rn CM ro
o o o o
i i i i
o- m n fx
ro ro ro ro
ro ro ro ro
i i i i
O O O 0
in -o -o N
o o o o
ro -o o- N
n CM n ro
-1 1H -0 0-
m -o -o -o
rt CM CM rO
O O 0 0
1 1 1 1
* n ro «r
CM ro •»• in
i i i i
-i CM CM ro
o o o o
o o o o
0 0 0 -i
o o o o
o o o o
o rx oo o-
O 0 0 O
o o o o
> to o ^
CM » -o rx
CM CM CM CM
~ n «r n
o o o o
o oo o
1 1 1 1
» 10 n o
o o o o
o o o o
n o n o
ro
CM
0 O O O
O O O O
o o o o
1 1 1 I
o o o o
o o o o
1 1 1 1
-o -o -o rx
0-0-0-0-
o o o o
t-i t \n -o
n in in n
o o o o
1 1 1 1
03 0- O *
-1 o -i -1
1 1 1 1
-1 CM CM M
O O O 0
IX 0- O M
0 0 <-"*
o •<> o n
ro ro * i-
CM 0- CO CM
N CO 0- O
rf ro t •<>
0000
1 1 1 1
o ro m rx
ro -o -a
i t t t
CM CM ro
ro «• n in
•0 O O 03
ro n -o -o
-i ro in ro
o o o o
1 1 1 1
CON fx »
PI » n rx
i i i i
-H n n T
o o o o
o o o o
0- o -i «H
O O 0 O
1 1 1 1
» oo CMIO
o o o o
1 1 1 1
mono
*4 — CM
rx
CM
249
-------
O
oo
a:
O
o
o
_i
o.
co
I
LU
_l
CO
-
0 00
O O O- CO
TH .H 0 0
111 O 111 O
•H rf CJ
CJ
rx -o -o TH
111 V TH CJ
N 0- T 0
P) CJ CJ CJ
111 CM -0 CM
P) P) CJ CJ
1 1 t 1
CJ 0- 111 TH
CJ TH TH TH
rx .H o co
111 O 111 O
.H TH CJ
CJ
O O -0 0-
o- rx cj o.
T -0 TH -0
fx CO *f CO
-o -o -o in
i i i i
CJ CO 111 O
CJ CJ CJ CJ
O -0 0- 0-
in o in o
O 0- 00 TH
O 111 111 CJ
0* PI PJ PI
T 111 111 111
1 1 1 1
O -0 CJ CJ
TH rx LI! eg
iii
TH 111 TH P-
«T CO PJ 03
co rx rx -o
O- O P3 -O
111 O U") O
TH .H CI
0-
250
-------
o
oo
a;
a.
to
a:
o
o
o
jj
a.
CQ
•=c
Ill
li. »
0
z
3
K
M
M
£L
on
SSOLVE
1 C
M U
a
o
a
z
o
HH
IRCQLAl
S B'
a. a
z
M
a
a
u
>
*TH
W 0)
U)
M
Q
pa
in
u. *
o
z
3
Ct
M
w
z
o rj
SSOLVE
a A
a
o
REGION
AU I
o o o o
o o o o
o o o o
1 1 1 1
o o o o
o o o o
CO 0} GO CO
» T in in
o o o o
CM N TH TH
o o o o
1 1 1 1
ro n -o C-
n ro n n
o o o o
i i i i
n •« in oo
00 TH TH
in to -o -o
»H N -o is
in co is o-
«r o TH -o
<-i M n in
CM Is Is 0 -O ^ *T
O «T 0- 03
TH (M n in
i i i i
in o bi o
TH TH (N
TH
0000
O O O O
O O 0 0
1 1 1 1
0000
o o o o
CO CO 00 CO
0- O 0* O
O TH O TH
o o o o
Ix fx -0 -0
o o o o
1 1 1 1
o- o -1 io
IN M M M
0000
1 1 1 1
o rx in
•O 0- -i M
(N rj co »
«n n «• co
o- n is -i
-0 CO 0- O
T-l
co M rf rs
IH o> ~o m
tin
i i i i
M in *o co
o o o o
o o o o
-< N M in
M M «• in
o o o o
--< in ^< rj
ro in co ^
O O O -1
in in co rj
o o -< n
-H -a n o-
rf « rj M
o o o o
1 1 1 1
O M -0 0-
-< rs » n
o o -< n
i i i i
in o in o
-> rt IN
»
O O O O
o o o o
o o o o
1 1 1 1
o o o o
o o o o
N rs rs is
0- 0- O- 0-
o o o o
o o o o
io ^ in is
in in in iii
o o o o
1 1 1 1
•0 0 IS
o o o o
o o o o
N >o *• n
r< n n r
o o o o
0- O 03 03
» in » v
o o o o
1 1 1 1
o » M n
T n n n
o o o o
1 1 1 1
>o in ^ is
o TH n in
•c •* -a n
M o n CM
•< n
^ Is O O
^ r^ *-< rH
•H n n
o TH rj N
M n * -o
T * -O 0-
C* » TH CJ
o TH n M
O 0 O O
1 1 1 1
n o- -o n
n CM co o-
o TH n *•
i i i i
n o n o
TH TH M
IS
O O O O
o o o o
o o o o
1 1 1 1
o o o o
o o o o
IS IS IS IS
TH TH O 0-
CN n n TH
o o o o
TH n M in
in n in in
o o o o
1 1 1 1
in n TH o
n in in n
n n ri n
i i i i
M in N o
O O O TH
(N 0- IS IS
TH TH oi n
«r n TH in
n -o is ix
» n -o TH
O V N O
a n M ro
CO 111 TH O
i i i i
00 -O TH TH
n in co o
i i i i
ri •»• n is
o o o o
o o o o
0- 03 CO IS
o -H CM n
o o o o
n oo TH iii
CM n in -o
0000
in co >o o>
O O O TH
oo n co M
O -H TH (N
o o o o
1 1 1 1
O IN TH IS
o n n 03
O O O 0
1 1 1
n o in o
TH TH [N
03
o o o o
o o o o
o o o o
o o o o
o o o o
Is rs Iv 03
o -H n t
•0 -0 -0 -C
o o o o
TH is >o in
M CJ M CM
o o o o
1 1 1 1
TH o n in
o n is n
o
M IS M M
TH 1
M *o in >o
N *o TH m
TH n M m
r< o o- TH
is o TH n
» Is 0- -0
TH
03 o is n
1 1 TH
1
•r 03 o- rs
•o .H in .H
1 1 1 TH
i
-------
O
CJ
CQ
in
u. o *o >o
o o o o
^- O rs in
* rf O O
O O O O
1 t 1 1
«r <«• -o o-
CN in rs o
0 TH
•H M CM M
n o in n
ro o -o ro
ro m >o co
rs o ^^ 1-4
.-1 * 0- in
1 1 1 r-t
1
-o in oo M
in o ^ o*
i i i i
O T-l » tO
O O O rf
o o o o
CNCN CO <0
•-> CN M -o
o o o o
O ••« IH 00
ro -< M in
VH CN ro »
o- -< o o-
» o in rf
*-i CN ro --o
a> * M M
O «i (M T
o o o o
1 1 1 1
CNO N rs
-0 0> CN rs
ro ^ rs O
CM M N N (N
oo o -o r^
O *-t ^ (N
00 0 r< O
si -a oo i
-< rf -i IN
*o o* v oo
O O lH rf
o o o o
1 1 1 1
in M in -o
O CN V fs
o o o o
1 1 1 1
mono
-H T-l CN
fN
v4
O O O O
o o o o
o o o o
CM CN rj CM
o o o o
o o o o
iii in -o -o
•0 >0 *0 *O
o o o o
rs -o in -o
o o o o
i i i i
t is rs co
n M n n
rs fs fs ts
M in rs o-
o o o o
0- M N O
O «H r* f J
ro M n m
rs o- «H M
1-( v^
rs >o ^ o
fs « o- in
»-l CI CN M
rs M •« o*
o * ** *->
i i i i
o- co >r CD
o in o in
i i i i
CN M in rs
o o o o
o o o o
rs CN en -o
O •<•< •* CN
o o o o
•H n o oo
«r in rs co
o o o o
in co rs co
^ «• ^o ch
o o o o
in oo CN rs
O O -H TH
o o o o
i i i i
rs » o- CN
0 CN » CO
O O O O
1 1 1 1
mono
T-l «H CJ
n
*-4
O O O O
o o o o
o o o o
CN CN CM CM
O O O O
o o o o
in in o •a
i-< *H r-l TH
O O O O
O. 0- 0- CD
O O O O
till
•« oo N n
m in in in
rs rs is is
M in N o
O O O rf
oo n rs o
CN rS rH 0-
co o ro m
o o o- oo
0- -0 CM O
rt M n
M M M ro
rj M < n
o o o o
O M -0 -1
rt in N N
>o rs co c>
rs oo -o CN
m * N IN
CN M n o ^
O <-< >< CM
o o o o
1 1 1 1
O CN CM <0
o CM n oo
o o o o
1 1 1
n o in o
•^ T* n
•o
T1
O O O O
o o o o
o o o o
i i i i
n n n n
o o o o
o o o o
>o N rs rs
*o in n n
IN CM CM T<
o o o o
0- O rf to
0 O O 0
o- is n «r
*H w fo *
o o o o
to >o o o
0* 0* ^ tO
O O ^ 1H
co to N n
O ^ *-t CM
o o o o
1 1 1 1
n » rs -e
o -< M -o
O O 0 0
1 1 1
n o in o
-4 ** r-i
IS
»•«
o o o o
o o o o
o o o o
n n r-i CM
o o o o
o o o o
n -o -o -o
o o o o
V t t t
o o o o
V t f t
r* r* r* r*
O O O O
1 1 1 1
oo o. ro -o
ro ro v *
CO 00 CO 00
CN ro to t
o o o o
O CO O- .-i
0 0 0 -<
00 0- 0- 0-
m -o N co
O CN CN 00
CO CN -0 0-
^ CM n CN
in CD o CN
0 0 -< rH
1 1 1 1
rs in ^ -o
•i n o- CM
i i i i
CM ro t n
o o o o
o o o o
•0 CN CO M
O ^ -* CN
o o o o
CO 0- CM *
CN n n >o
o o o o
rs N n to
-o rs o CN
o o o ^*
•o o- -< «r
o o •< -«
o o oo
i i i i
» o ro rs
1 » N O
0 0 O^
1 1 1 1
mono
x xf-1
CO
252
-------
in
U-f
o
2
3
cc
n
t-H
a.
QN
[SSOLVE
:i (
a
Q
ffi
2
p
M
:RCOUA:
5 B-
a. o
2;
ca
p
UJ
_i
U) A
cn
M
a
CQ
llT
U. »
0
Z
3
cc
n
M
Z
a CM
SSOLV
1
a
o
z
0 3
1-1 «
O
UJ
0£
O O O O
o o o o
o o o o
o o o o
0000
in -o -o -o
o o o o
w n M m
o o o o
in m in -o
n N N N
o o o o
i > i i
n is TH n
TH TH CM (M
rs rs rs rs
1 1 1 1
ro ^ in *o
0000
03 TH to -43
O TH TH TH
rs TH «r rs
MJ 00 0- O
TH
N IS 1H TH
oo N rs .H
TH
o o o o
r> « n n
(N N M tN
0000
1 1 I 1
n ^
o -i .H (M
0000
o n oo N
i1 in -o oo
o o o o
•H V N rs
oo oo o n
O O -1 -1
lf> OK f) OK
O O -> TH
o oo o
1 1 1 1
tr n o rs
o CM m a)
o o o o
1 1 1 1
in o m o
TH -I CM
o
r»
o o o o
o o o o
o o o o
o o o o
o o o o
in in -o -o
00 03 CO 00
r*i M n ro
o o o o
(N TH M N
M N M N
O O O O
1 1 1 1
«r -o n TH
«• o in
OK N «• fS
i i i
M n in rs
o o o o
o o o o
in rf m -o
O T-I TH M
o o o o
M in o- «•
t 111 -0 CO
o o o o
"< 0- 00 O
rs ^) rs o
0 0 O rf
in o- n oo
O O -H -<
o o o o
1 1 1 1
n oo o- in
O *H M »0
o o o o
1 1 1 1
in o in o
-H ^ ri
r4
M
O O O O
O O O O
o o o o
1 1
o o o o
o o o o
•o -o •« rs
in t T o >o >o >o
M n CM n
M CM (M CM
o o o o
•o N co ci-
io n ro to
o o o o
1 1 1 1
CM OK OK T-I
o
-------
o
oo
r-J
a:
o
o
_i
Q.
LCI
CO
«=E
in
u. *•
o
z
D
CK
ro
M
0.
cm
SSOLVE
1 C
Q
O
CO
z
o
M
„ \
CO CQ
cn
n
o
A
«
U. «t
O
z
gro
TH
2
O CM
SSOLVE
u e
a
o
REGION
AW f
O O O O ,
o o o o
o o o o
1 1 1 1
o o o o
o o o o
o o o o
1 1 1
CD 09 CO CO
in in i/> in
o o o o
o o o o
CM PI TH -H
O O O O
1 1 1 1
ro ro -o o*
to ro ro ro
0000
1 1 1 1
to -o in oo
0 0 -H TH
in o TH o
TH ro IN o-
THfN -0 0-
* O TH -0
TH CM lO in
TH TH 0- rN
•0 t IN IN
TH PI
mono
O TH PI PI
i i i i
N IN CO -0
in o- -o CM
TH CM «T CD
i i i i
IO «• IN O
O O O TH
O O O O
-i -i o O
o o o o
1 1 1
0- 0- l«l 00
rs in m IN
O TH PI IO
CM TH o n
o- ro «• IN
CO « -0 TH
TH TH CM
«• IN pj m
O O TH TH
o o o o
i i i i
t V 00-
» -o ro T
o TH ro in
i i i i
n o in o
TH TH CM
TH
O O O O
o o o o
o o o o
1 1 1 1
o o o o
o o o o
o o o o
1 1 1
00 00 00 00
•o -o -o IN
o o o o
o o o o
IN IN -o -o
o o o o
lilt
0* O TH rO
CM ro 10 ro
o o o o
1 1 t 1
» Is *o rs ^~
n -o o in
o o •< -<
-H CM <»• rf
CM CO 00 0>
CO 0* ^4 tO
*H rH
•0 O » 0
O iH *^ *H
o o o o
1 1 1 1
oo o o <-i
-< IN m in
O O ^ CM
1 1 1 1
in o in o
-I -I IN
CM
O O O O
o o o o
0000
1 1 1 1
o o o o
o o o o
o o o o
1 1 1 1
IN IN IN 00
-0-0-0-0
o o o o
o o o o
«?• in -o IN
to to to to
o o o o
1 1 1 1
to co rs -o
in
«H r-(
o> o f T 00 SJ
in oo o -o
oo -i in -o
r-l »H r-t
in ro Pr
-------
O
o
LT)
I
CO
n
u.
CM n rs o
«r t «r in
i i i i
oo oo o- in
O rH tO -0
03 rs rH tO
-IM o- in
rs «r in o
-oo- -H «r
rH rH
t oo rs n
O O rH 10
o o o o
1 1 1 1
f rH 00 -0
o rn N n
1 1 1 1
no n o
rH rH CM
O
0 0 O O
0 0 O O
o o o o
0 0 O O
o o o o
1 1 1 1
•0 -0 N N
oo rs TS rs
ro to to ro
o o o o
•o «r «r «r
CM CM CM CM
o o o o
1 1 1 1
CM «r o n
rH rH rH O
O O O O
1 1 1 1
in o N ro
O rH rH tO
rH oo in t
CM 00 00 O
o «o- n
rH rH rH CM
oo o- in oo
rH ts Hf CM
n to rs o>
o rn CM in
1 1 1 1
o- rs rs o-
t rH CM ro
i i i
rH CM n o-
o o o o
o o o o
rH CM CM 00
O O O O
1 1 1 1
oo o n N
00 CM -C CM
O rHrH CM
tO O n rH
ro «r n rs
i i
» •• in 0-
f t ro n
in in n in
o
o o o o
o fi in o
o M rn n
O 0- 03 03
o -o
ro ro ro to
0 O O O
««*«
O O O O
III!
oo o- ro -o
ro ro T t
rj ro «r in
o o o o
ro o^ o D-
ri ri ro ro
r-i w \nr**
o o o o
O O O 0
lilt
O' *H N *
o o o o
III)
in o in o
00
255
-------
O
o
LO
UJ
_l
CQ
in
U-
_i
o
01
CO rt
a
o
CQ
o
IRCOLA1
5 B'
0_ CQ
Z
a
Ul
en n
tn
M
a
in
u. «•
o
z
cc
z
QCM
SSOLV
U
o
o
REGION
AU t
O O O 0
o o o o
o o o o
o o o o
o o o o
1 1 1 1
in -o -o -o
03 03 00 00
CM n CM n
o o o o
m in in -o
cj CM n ci
o o o o
1 1 1 1
fM N rt rO
rt rt CJ CJ
rs rs N rs
i t i i
CM «• in -o
o o o o
is o- rt rs
O O rt rt
oo rt in o
in rs oo o
rt
ro o o CM
•0 O. rt CM
ro -o o- CM
OO O rt
1 1 1 1
ro i- -o r rt o-
cj ro r
o o o o
IIII
in o m o
rs
CI
256
-------
oo
o
oo
GO
o
a:
o
o
o
Q_
C3
•z.
I—I
o;
CQ
V
v
3
CD 00 N
(M CM CM CM
n o n o
1^ CM
-
N » -o n
N » 09 rf
09 09 O n
n ^ V M
tn -H o ^
•a n«r M
(N CMfM CM
i i i i
N n OK i-*
N •<> n »
» » T »
1 1 1 1
oo -IM n
•o -o n *
CM CM CM CM
n on o
«•• ** CM
CM
N n O n
OK N o n
» -OrH -0
N -o -o n
in * t N
O- OK 00 N
MMMM
1 1 1 1
M O O CM
rt M » »
n n n n
i i i i
CO OOO M
OK O OK OK
M * M M
n o n o
rH rt CM
CM
M oo -o n
00 •* N 0
n m N M
rs -o in m
09 00 * -0
CM CM CM rt
MMMM
1 1 1 1
TH CMN CM
N OKO CM
» « n n
1 1 1 1
X N OK N
MMMM
MMMM
n on o
*1 •* CM
CM
V CM O CM
-1 •«•» N
M N » rt
N -0 -O -0
n oo co N
V <• » »
MMMM
1 1 1 1
•o •" rs N
O MM •»
n n n n
i i i i
n OK 00 N
MMMM
n on o
** ~* CM
M
M -0 CM 03
» OK n •«
•0 O N*
N N *Q -0
O •< -1^
n n nn
MMMM
1 1 1 1
00 O M CM
O M M »
-0-0-0-0
1 1 1 1
O CM CM -1
n n nn
MMMM
n o n o
* ** CM
CM
CM
CM -0 * rt
» N 00 "
in -o o-o
rs -o -o m
00 N CM *
N -o n »
MMMM
1 1 1 1
-H o n-o
N N N 03
MMMM
till
CM rt N «•
CO 00 NN
MMMM
n o no
~t * CM
M
iH
00 M rf 00
M OK CM O
» t 03 M
-o n *• »
09 <• » »
CO CD N N
CM CM CM CM
1 1 1 1
~ -O CMN
O CM MM
n n nn
i i i i
* » o -e
OK OK OK 03
CM CM CM CM
n o no
-H n H
1 1 1 1
09 ¥5 CM -0
n 03 o- co
1 1 1 1
00 M 09 N
M » M M
CM CM CM CM
n o n o
r* 14 CM
N
CM 0 » N
M -< «r N
M n o -o
rs -o -o in
o <• N N
09 09 03 09
MMMM
1 1 1 1
CM -0 CM »
•0 OK N CM
niffo-o
1 1 1 1
CM OK CM rt
CO 09 OK OK
MMMM
n on o
-« rH CM
•0
CM
n ^ -o «•
rH CM CM OK
n -o ^ -o
N ~c ~e n
rt OK -0 «•
» M M M
MMMM
1 1 1 1
» -0 M 00
O CM M M
-0-0-0-0
1 1 1 1
CM » CM *H
MMMM
mono
-< TH CM
N
» O N M
09 OK * M
OK « -0 CM
-o -o in in
N 09 n »
CM CM CM CM
MMMM
1 1 1 1
<• T* 09 »
n 09 OK rH
n n n -o
i i i i
T* 00 OK CO
MMMM
MMMM
n o n o
-t .H CM
09
•O N 09 OK
-0 N 1 09
N 03 M 00
•o nn v
CM » M CM
•fl -0 -0 -0
MMMM
1 1 1 1
t cMn n
M -0 OK *
in in in -o
i i i i
» OKM n
•0 -0 N N
MMMM
n o n o
•1 * CM
N
CM
O O OK n
o- 1 n n
03 CM N M
N N -0 <0
OK » 00 n
OK 0« 00 09
MMMM
1 1 1 1
CM M M 00
09 OK OK OK
-0-0-0-0
1 1 1 1
OK 09 M O
OK OK OK OK
MMMM
n on o
rt rt CM
CO
oo n n CM
^ * OK *
09 CMM N
QK N •«
rt O OK IS
» » M M
1 1 1 1
•0 CM M CM
rt O OK 09
CM CM ** •<
n o n o
rt -• CM
OK
257
-------
oo
i
i—i
O
<=c
M
UJ
o:
o_
o;
o
o
o
_i
Q.
o:
Q-
OO
LLJ
_J
CO
<:
m
u. *
o
§
cc.
n
M
0.
Q N
tSSOLV
;i
Q
O
a
z
Q
M
4 0)
-1
o
u
ot
a. a
z
w
IB
s
3 ,
en oo
tn
M
a
A
n
u. »
r
i
n
M
z
a CM
SSOLVE
»1 (
a
o
REGION
AW t
00 O O
o o o o
OO 0 0
o o o o
OO 0 0
1 1 1 1
•o-o r\ rs
HIM N N
OO O O
l"t rH rH rH
O-O O- 00
rH CM i* rH
OO O O
1 1 1 1
O » rH •H CM
i i i i
co xi o r-
CMn o> w
rH rH rH CM
1 1 1 1
is n » »
o •* CM n
oo o o
TH 00 QK SI
(MO » 0-
OO 0 O
1
rs rH rs rH
CM CM rH CM
o o o o
(NO n CM
rs rs n <
inn n in
•« -i «r o
rHM «• -O
o o o o
1 1 1 1
-0 03 O •«
rH CM M CM
O O O O
n o n o
•1 X (M
«H
o o o o
o o o o
O 00 0
o o o o
o o o o
is rs rs rs
OK in o, oo
» » n n
o o o o
*• » -o in
N CM (M CM
O O O O
1 1 1 1
•O N CM 00
rH O rH O
0 OO O
1 1
in rs o *
0 OrH rH
•Q co in »•
CM n in is
«• CM o- n
n -c ~a is
oo CM o n
oo M rs OK
CM M M n
OB CM is n
O rH rH CM
i i i i
IO O CM rH
o- r< CM n
1 rH rH rH
1 1 1
n oo CM n
o OrH *
o o o o
n o M n
0 CMT -0
0 OO 0
in o o rH
rH CM CM »
o o o o
» fs OK CM
«• -ooo in
CM CM CM n
-< 0- » O
-i »< CM n
o o o o
1 1 1 1
CM -o o in
o -< <• -o
o o o o
i i i i
mono
^ o -o in M
»H T-l TH rt
rs ^ -o -<
O rH rt CM
O O O O
1 1 1 1
» x » 00
o o o o
o o o o
1
n o n o
-i -< N
n
o o o o
o o o o
o o o o
oo o o
o o o o
i i i i
in in in -o
o o 03 -o
0- O- 00 00
o o o o
in o- o CM
n M <• <•
o o o o
1 1 1 1
M < » n
-0 CM CM r*
CM CM CM CM
1 1 1 1
v -o rs o
o o o -H
oo rs in rs
•rf CM M 0 O M
O O X 1
O O O O
CM o in *
^ o ^H ro
0 0 O 0
1
CM O Cf. »•
o o o o
rH CD CM O
CM CM CM CM
n m nn
OK rs OD O
O rH CM *
O O O O
1 1 1 1
N -0 » *
rH O O rH
O O O O
1 1
in o in o
rH rH CM
*
o oo o
o o o o
o o o o
o o o o
o oo o
in in -o -o
in CM o- oo
in in T OK »
O rH -H CM
1 1 1 1
-0 O- O (N
rS 000 rH
1 1 rH rH
1 1
M nrs o
O O O rH
o o o o
<«• -O 00 rH
O O rH rt
O O O O
1
rH rH O rH
O O O O
< rH oo rs
oo 03 in »
r-l rH T* rH
oo in CM o
O rH CM M
o o o o
1 1 1 1
M » O rH
O O rH rH
O O O O
1
in on o
rHrH CM
in
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 1
-o rs rs rs
o- in n n
0000
03 o- n o
rt m *o rs
o o o o
1 1 1 1
03 T CM 0>
rs in o rn
M rH rH O
1 1 1 1
•o o co n
O rH rH M
n n m o-
CM * rs n
rH
0. » rH 0
o- M rs o
rH rH CM
N n N -o
co n oo n
M in -o oo
in o- •< *
rn CM in o-
1 1 1 1
•o in CM •»
o. o CM n
rn n T in
i i i i
•o » n m
O — M n
o o o o
rH CM O O
CM rH CM -0
O 0 0 O
1 1
n 03 O rH
hi CM M V
o o o o
CM O -0 0-
T o- in »
in -o rs co
M in CM -0
i n-oo-
o o o o
i i i i
n rs o- o
CM rH CM CM
o o o o
i
in o in o
rH rH CM
•0
0000
o o o o
0000
o o o o
0 O O O
1 1 1 1
in -o -o -o
o> rs in ro
o o o o
m CM rs CD
M V ^ ^~
o o o o
i i i i
rH rH » 03
o n o- co
in » n n
i i i i
n o- «• n
O O rH CM
03 n rH CM
rn M in oo
m «• rH in
•0 00 0 rH
rH rH
» CM O 0>
CM 03 T 00
CM CM M M
rH CM M -0
.H CM M in
1 1 1 1
O 10 rH -0
rH n o «r
rH rH CM TJ
1 1 1 1
in o> » -H
O O rH M
O 0 O 0
N M CM *O
-H O rH M
O O O O
1 1
-0 O O M
n «r in -o
O 0 O O
in CM rs rs
ro co n o
n n * «r
OD o^ ro «-*
o -i M m
o o o o
1 1 1 1
rH 00 o o o
O rH rH rH
00 rH <0 -0
rH 0 rH rH
O O O O
1
•o M n rs
rs -o » o-
N 0* rH rH
rH rH CM CM
1 1 1 1
0 O -0 rH
rH CM M »
»• in o CM
CM in o CM
rH rH
rH » rs 0-
n oo CM o
rH rH M n
noon
* O- CM »
n t ~o a>
» 00 03 rH
CM » 030
1 1 1 rH
1
o in n rn
-o o CM <•
CM » in rs
i i i i
CO » *• rH
O rH CM V
0 O O 0
» -0 rH CM
O CM -0 rH
O O O rH
rs in n »
* -0 O- rH
O O O rH
03 00 n CM
rs in -o o
CM CM CM M
oo in on
rH M -0 O
O O O rH
1 1 1 1
» rs in »
» -o CM rs
O O rH rH
1 1 1 1
n o in o
rH rH CM
0-
258
-------
UJ
_i
OQ
i oooo
m i oooo
1 OOOO
i
b. « 1 OO OO
O 1 OOOO
Z 1
oc i »nn-o
i
CL 1
i MM nn
Q N 1 »OK 0-0-
:> 100 00
ei !
in i 03* oo-
at * \ MM M-«
Q IOO OO
i i i i i
i
i -o-o noo
o i nrs Ttn
i inn -o-o
i M* rf»n
i i i i i
i
i on «Hfs
A 1 O-< MM
i
Z 1
O 1
M 1 03 M N t-o
t *4 »«»*
Z 1
M 1
i nfs nM
A i r*rx no
UJ I
s i
_i i moo •«-<
U) A 1 <*M »00
(n i i i i i
t-4 1
a i
I o-x 00-0
« 1 nM fsM
1 r*M nn
1 OO OO
Z i ' ' ' '
03 I mono
1-4 4 1 r4 *H N
O 1
Ul 1 O
tC 1 "1
OOOO
oooo
oooo
oooo
oooo
n » » n
o- o- CD m
fs IS rs rs
oooo
o- -in in
4) tsrs Is
OOOO
1 1 1 1
rs «r fs oo
n 0- » n
oo rs. rs is
i i i i
ID CD-I IT)
O O rt Tt
M 00 » n
rt rt (M M
•0 O t-l N
•0 00 0- O
«H
M M 00 »
n •« 03 .
1 14 »4 T4
i i i
n o- n co
0 0 rt ,-.
oooo
0- « 0- »
O O O M
oooo
1 1
M -0 00 M
n n n »
oooo
o- -«n o-
M fOM •<
M M M (N
o rv •« o-
*4 IH N ro
oooo
i i i i
N n N -o
oooo
oooo
1 1
n o n o
«
oooo
oooo
n o in o
000-
oooo
1
O rt M -H
M M M M
OOOO
rs o o- rs
o — o o
n 09 M 41
O O 03
n « M
M CM N M
^ 0" » O-
rs N 00 00
•« o- •< n
O O — •"<
1 1 1 1
n o- -o n
» — n in
«i M CM M
oooo
» » rs N
n » n M
» co M n
o o «-i •«
oooo
1 1 1 1
in rs r< -o
-i rf M M
oooo
1 1 1 1
n o n o
rf — N
CO
259
-------
ca
n
u. »
0
or
m
M
a.
a CM
[SSOLU
:i
a
o
IB
Z
0
:RCOLA
t B
a. a
z
a
a
Ul
^
j
in n
in
M
Q
a
n
u. »
o
z
m
M
Z
QCM
Ul
_J
O
to
in TH
a
o
REGION
AU (
OO O O
o o o o
oo o o
CM TH TH TH
oo o o
o o o o
1 1 1 1
OO TH CM
10 CM TH OK
rsrs rs o rs o.
o o o o
o HT oo CM
CM *r •« rs
n m mm
o n oo CM
rs rs rs co
O TH TH TH
1 1 1 1
rs CM » on
m <• w
o o o o
1 1 1 1
ro TH n rs
o o o o
1 1 1 1
n o n o
TH TH CM
CM
o o o o
o o o o
o o o o
CM TH TH TH
O OO O
O O O O
1 1 1 1
0 TH TH CM
TH TH TH CM
00 00 00 00
O O O O
CM » n OD
o o o o
0 O O 0
o m v o-
M V N O
CM CM CM h)
1 1 1 1
m t n -o
o o o o
O M -0 OK
N OK O TH
CM CM m m
O -O TH <•
00 00 OK 0-
rs OK CM n
0 0 TH TH
1 1 1 1
o -o o »
n n -o -o
i i i i
CM m » -o
o o o o
o o o o
n CM ro o
0 0 0 TH
o o o o
i i
CM n CD TH
CM CM CM M
O O O O
n n n o
+ n n -o
«• oo TH n
O O TH TH
OO O 0
1 t 1 1
00 <• CM CM
TH CM m »
0000
1 1 1 1
n o n o
TH TH CM
n
CM
TH O TH O
o o o o
o o o o
CM M CM CM
O O O O
0000
1 1 1 1
00 OK 0 0
n n n n
0 00 O
ro CM M »
CM CM CM CM
o o o o
OD ro n n
•o in oo o
n n n -a
i i i i
t -0 CO O
O O O TH
o n o- »
CM » n v
m mm m
<• -0 -0
o o o o
n rs CM CM
-1 TH CM CM
o o o o
1 1 1 1
n -o n n
•o in TH ro
CM CM CMM
1 1 1 1
•o o * o>
0 TH TH TH
ro -H o o
TH CM ro »
•0 N -0 TH
» -0 10 -fl
oo CD oo rs
co n ro TH
o TH CM ro
i i i i
rs o o N
» n T «•
i i i i
«• -o o. CM
O O O TH
o o o o
ro o -o »
O O O TH
o o o o
OK » -CIS
CM ro ro to
o o o o
O TH O. 00
m «r *
-------
oo
i
p-H
O
oo
CD
O
•ZL
_i
en a
U)
M
a
m
Hi
u. *
a
z
3
ac
to
z
m «• n -o
•O TH -0 H>
o TH •< (N
I I I i
to ro -o co
-0 O -0 N
1 TH TH (M
1 1 1
-at m v
o TH CM ro
o o o o
ro -0 ~0 03
-o oo TH in
O O TH -1
a> TH rs o
CM CM TH CM
o o o o
ro in o CN
-o » n TH
o o o o
1 1 1 1
rs o- in -o
o TH ro in
o o o o
1 1 1 1
v CN N in
O CM M in
CM ro «r -o
o o o o
in co •« CD
*H TH CM ro
o o o o
in o. is -o
M 00
O TH CM CM
o o o o
1 1 1 1
-i n CM -o
o o o TH
o o o o
1 1
in o m o
TH -< CM
CM
o o o o
O O 0 O
o o o o
o o o o
o o o o
in in -o ~o
o o o o
o o o o
1 t 1
-0 0- O TH
M IO «• ^~
o o o o
1 1 1 1
O. O O CM
CM n ~ O
iH »H TH *H
1 1 1 1
CM M M M
o o o o
-1 T -O CO
in -o oo CK
-o CM n f
t in in in
rH TH in i-l
^- •« co o
O O O *i
1 1 1 1
O CM -0 -H
rf rf rf CM
1 1 1 1
tn M o o* c-j *r
0 O -H TH
CN in CD CM
O O O TH
o o o o
1 1 1 1
CN -H -0 -1
O O O TH
o o o o
1
in o in o
TH TH CN
in
o o o o
O 0 O O
o o o o
o o o o
o o o o
*0 N IS N
o o o o
o o o o
1 1 1 1
co o- no
n in -o is
o o o o
i i i i
00 t CM 0>
is in o -H
ro TH TH o
i i i i
in o N to
o TH TH n
CD Is CD O
TH M -0 W
in CM o is
CD CM S) 0-
TH TH TH
•H tO CM CM
co oo -H in
to ^ >o rs
CD O IO CD
O CM f 00
1 1 1 1
CM O IS *
co o in -o
i TH n to
i i i
in CM CM to
O TH CM IO
o o o o
TH IS O* O
in -o CD CM
O O O -H
to CD 03 O
0 0 O 0
*o ^- co to
~o TH «r n
O O O TH
1 1
*0 CO &• "0
o -i to -o
o o o o
1 1 1 1
is -o ^- in
-H O CM CM
o o o o
1
in o in o
TH TH CN
-0
o o o o
o o o o
o o o o
o o o o
o o o o
in -o •« -o
o o o o
o o o o
in CM rs co
to ^- •» v
o o o o
1 1 1 1
.H .H rs in
.H O O O
o o o o
1 1 1 I
in o in o
TH TH CM
03
o o o o
o o o o
o o o o
o o o o
o o o o
1 1 1 1
in in -o -o
rj to to to
o o o o
03 .H O -0
.H 0 -H rt
o o o o
1
-0 n to rs
IS -0 » O-
rs Os TH TH
TH TH CM CN
1 1 1 1
o o in o-
TH CN fO fO
i- N «r oo
CM «r o- o
CM 0- 00 TH
TH -0 -H O
TH TH fN fO
O -0 O CN
-o in in CD
to i- in is
•0 0- IS -0
TH ro rs oo
i i i i
00 03 CN O
•H CN 10 -0
1 1 1 1
rs n to o
o TH CM
-------
o
o
00
I
CQ
1
in i
i
i
i
u. «• i
a >
Z 1
D 1
a i
ro i
" 1
1
LL 1
1
a CN i
:> i
_i i
a i
CO 1
en x i
a i
i
i
O 1
f
i
i
i
i
SO 1
1
Z 1
O 1
K- * 1
RCOLA
B
a. 03 i
i
Z 1
t-t t
i
a i
U 1
^ i
j i
en a> i
en i
M t
a i
i
dO 1
i
i
i
i
i
i
in i
i
i
i
u. o CK in o
1 1 -1 CN
i i
in o N oo
O x rH CN
oooo
IH co CK ro
oooo
CK CK >r ro
oooo
ro *o co **
~< m *c CK
TH TH X TH
in CN in rs
o ^ ™ «•
oooo
1 1 1 1
N 00 -0 O
X X 1 (N
OOOO
1 1 1 1
in o in o
X IK (N
O
OOOO
oooo
oooo
oooo
oooo
ro «f ^ iii
0- 0- 0- O
-H rt X CN
oooo
o- ^ in in
>o rs rs rs
oooo
1 1 1 1
rs T rs co
ro CK T in
CO N rs is
i i i i
in co -< in
0 0 « -i
o t oo is
TH rH « CN
00 O O CN
m rs co CK
rt CD rs -o
in rs CK o
incK-o -o
O O x (N
1 1 1 1
in >o rs in
ro r o- in
oooo
CN 0 x o
O O O rf
oooo
1 1
in o m o
oooo
oooo
oooo
oooo
oooo
CN ro ro ro
in in in in
oooo
oooo
O CD CK O
ro CN CN ro
oooo
1 1 1 1
ro T-I co -o
in rs N rs
in in in in
i i i i
ro
i i i
ro * >o
ooo
ooo
ooo
0- CN CM
ooo
CD CK in
o
ooo
ro ro ro
x ro CM
N rs rs
•o en CM
O O -1
i i i
^ CN O
1 1 1
•T •« CO
OOO
ooo
ooo
M -1 O
ooo
to rs oo
ro ro ro
000
o o -<
ooo
1 1 1
x CM rs
ooo
ooo
1
o in o
X »-» CN
oooo
oooo
oooo
oooo
oooo
i
oooo
CM ~o rs -o
in in in in
oooo
III!
in CN rs o
rt CO CO O
ro CM N ro
i i i i
CN ro
-------
O
o
00
I
in
u. *
0
z
D
a.
n
tH
o.
SSQLVE
:i c
Q
o
n
z
o
M
h- t
RCOLA
B
0. «
Z
n
«
Ul
:>
-I
co n
(0
M
a
a
*!
u. <•
U. 1
o
z
o
n
M
Z
a M
1 T*
jmoss
a
o
REGION
AW t
o o o o
o o o o
o o o o
•1 M (M (N
O O O O
00 O 0
1 1 1 1
O O « M
o o o o
NX » «
CMM n n
oo o o
co is ^ m
no » in
rco o* -o
W n *0 IS
IH *H TH TH
(N M in 00
o o o o
o o o o
1 1 1 1
o o o o
*H vi «4 iH
oo o o
1 1 1 1
mo in o
•H •< (N
0-
H
O O O O
o o o o
o o o o
o o o o
o o o o
0000
1 1 1 1
i ** CM CM
o o o o
0- O M M
-I CM CM CM
O O O O
1 1 1 1
*r o- IT n
-0 * (N (M
is rs rs rs
«r in -o m
o o o o
0- -1 M -0
M M M rO
•H o in o
CK 0- 0 n
o
»•( 1-1 TH »*
o o o o
1 1 1 1
o oo o- in
n CM o o-
ix is rv •«
T *0 N 0-
o o o o
oo rf « rs
CM M M n
N (M >0 ID
fx CO 00 CO
» sO 0- M
O O 0 -1
1 1 1 1
in in t CM
i i i i
m ^ >o oo
o o o o
O O 0 O
O ^ M in
o o o o
CM CM Cv| CM
O O O O
oo *• co in
in -o -o rs
o o o o
CM 5T rs O
o o o -<
O O O 0
1 1 1 1
«»• n -i CM
o o o o
o o o o
1 1 1
n o in o
rt rt CM
*H
CM
O O O O
0 0 O O
O 0 O O
o o o o
o o o o
o o o o
nn n t
o o o o
If) »• CD 00
M M M ro
o o o o
till
in 1-1 o o
o rs oo o
-o in in *o
rf CM CM W
o o o o
03 O rt CM
n o~ M in
rH rt CM CM
-< m » >o
o o o o
1 1 1 1
IS -0 -0 N
O O O O
1 1 1 1
r-t -H CM M
O O O O
O O O 0
in in >o is
o o o o
-1 rf -. M
o o o o
-< * 0- O
o -i CM in
o -< -i «
TH CM n t
o o o o
o o o o
1 1 1 1
-o in » n
o o o o
0000
1 1 1 1
in o in o
-1 •< CM
CM
CM
0000
o o o o
o o o o
o o o o
o o o o
o o o o
M M « m
n n n t
o o o o
00 •-< CM CM
M o rs N
o o o o
o is -o in
CM CM CM CM
o o o o
1 1 1 1
o o CM n
M 00 O CM
oo rs co 03
CM n n t
o o o o
o CM «• in
— < TH 1-4 T-l
t^ V CO ^
-1 CM N n
CM t ~0 0-
o o o o
1 1 1 1
00 00 CC 0*
o o o o
1 1 1 1
** CM n «•
o o o o
o o o o
rv co o* o
o o o o
-1 -,-,-.
o o o o
r-i CM 00 M
a- rn CM *r
O -I -1 -<
-H M in rs
o o o o
o o o o
1 1 1 1
•O *• »H ^H
0 O O O
0 0 0 O
1 1 1
in o in o
« rf CM
»
CM
O O O O
o o o o
o o o o
o o o o
o o o o
0000
O r* n CM
in in in in
o o o o
CM t 111 CO
O O O O
o o o o
o M
O O O O
1 1 1 1
-< o o- rs
-i -< O O
i i i i
-1 CM M in
o o o o
o o o o
CM * in N
o o o o
CM CM CM CM
O O O O
CM o- co rs
•o -o rs co
o o o o
« (N » -0
O O O O
O O 0 O
1 1 1 1
rs oo oo oo
O O 0 O
o o o o
1 1 1 1
in o in o
-« -1 CM
in
CM
-10^0
O O O 0
o o o o
o o o o
o o o o
o o o o
03 0- O O
n 1 1 T
o o o o
M CM n »
CM CM CM CM
O O O O
co M in m
-o in co o
in in in o
i i i i
«• •« co o
O O O -f
c<- n -o o
» O 0- CK
CM M CM N
-< in in M
00 00 00 CO
M -o o in
O O ^ ~!
i i i i
•0 IS. V «
«-l «-l 1H «H
1 1 1 1
CM T in rs
o o o o
o o o o
o ••< n -o
o o o o
CM CM CM CM
O O O O
r< CM -o GO
M ^ ^ ^~
o o o o
CM n -o ct-
o o o o
o o o o
1 1 1 1
CM * M CM
o o o o
1 1 1 1
in om o
rf -H CM
•0
CM
O O O O
O O O O
o o o o
«H iH *-l *H
o o o o
o o o o
-< T-l IN M
in in » «•
o o o o
in is CM CM
rf -1 CM CM
O O O O
1 1 1 1
in «o in in
-o in -I m
CM CM CM CM
i i i i
•0 O » 0-
O rf "i rf
rf -0 » N
CM n CM oo
*• v «• n
CM in in rs
0- 0- 00 -0
in o co oo
O -1 ** CM
i i i i
-1 O N M
CM CM *H -4
1 1 1 I
^ -0 0- N
O O O rf
O O O O
CM < 00 M
O O O O
CM M n M
o o o o
o- «r •« »
n M fn *•
o o o o
M in o •«
O O « -I
o o o o
1 1 1 1
n T CM rs
0 0 OO
1 1 1 1
in o in o
-< t CM
IS
CM
263
-------
00
o
oo
ce:
o
o
I
0.
C.D
Z
i—i
o:
oo
CTl
I
„
3
0 >0 GO 1*1
in » M n
in VH o -H
-0 I/) » M
CM CM CM CM
1 1 1 1
-< 0" >r o
n TH (N CM
CM CM CM CM
1 1 1 1
ao -< n in
•Q *e \n T
CM CM CM CM
in o in o
CM
~0 -O iH CM
•T CM -0 CM
CM * CD »
rs >o in in
in -< * rs
o- o- 03 rs
nm m m
i i i i
CM in o -o
-0 00 O O
(N CM n n
i i i i
oo o oo n
0. o 0- 0-
n t nm
in o in o
-H T-I CM
TH
CM
CM -O rH O
< T* in o-
00 O » 0-
•o -o in n M
n o in o
n
«r TH in ro
n co in N
is -o -a -a
O TH TH O
in in in in
ro M ro ro
i i i i
o- in in ro
in oo o- TH
ro ro ro o. in
ro CM TH o
o o o o
CM TH N HT
oo a) is rs
ro ro ro ro
in o in o
TH TH CM
ro
CM 00 CM -0
N oo ro 10
TH TH m o
-o in v r 03
CM CM CM CM
in o n o
«r
IS -H «f TH
TH rs •« O
00 00 03 03
ro ro ro ro
in o in o
TH TH CM
t
03 00 &> CO
•O 0- 03 •«
in rs CM co
-o in in T
TH co is in
M TH TH TH
ro ro ro ro
I I i I
o- -o o to
o CM in -o
M ro ro ro
i i i i
ro in «r ro
CM CM CM CM
ro ro ro ro
in o in o
in
03 -o
•o oo -H in
o- TH rs ro
-o -o in in
O 00 IS -0
0- 00 00 03
ro ro ro ro
i i i i
-------
oo
o
oo
a
a
o
o
a:
a.
a:
o
o
_i
Q.
a:
o_
oo
o
C\J
i
ca
<:
in
u.*
|u
cc
n
rH
CL
a CM
SSOLVE
1 C
O
o
m
at
z
o
rH
1 <:
o
0
si
a
UJ
o o o o
oo o o
oo o o
Tin in m
oo o o
o o o o
1 1 1 1
-o -o N rs
rH rH rH rH
rH rH rH rH
O O O O
CM CM N CM
OO O 0
1 1 1 1
(Nin T CM
o o o o
in 03 T 00
OO -H CM
CM in TH »r
CMM -0 CM
rH
O CM -0 T
O- rH n -0
rH rH rH
M to T in
OO- O CM
T «r -o is
rs rH O O
O TH CM T
1 1 1 1
in oo «r is
n -o o T
rH rH CM CM
i i i i
•On -H rH
o rn CM n
o o o o
OTTO
M O M 00
o o o o
1 1
T 0- -0 Is
M CM CM CM
o o o o
o- o TH rs
IS O O- O
is co rs oo
in 03 o in
rH CM T in
o o o o
1 1 1 1
in rH 0- O
CM T T in
o o o o
in o in o
rH TH CM
TH
o o o o
O 0 0 O
O 0 O O
o o o o
o o o o
i i i i
MJ -0 N rv
CM CM O O
0-0-0-0-
0 0 0 O
T T in in
CM CM CM CM
o o o o
1 1 1 1
CM -0 T rH
O O O O
in CD CM oo
O O rH rH
CM T -0 0.
in n o- 1*1
T rv o- o
ho n ho T
rn -o in CM
in -a is oo
rH -0 CM 03
rH TH CM CM
1 1 1 1
o- n «r T
rv o- o -H
1 1 -H TH
i i
CM in rs o-
o o o o
o o o o
T in T O
CM rH O rH
o o o o
1 1 1
TH o oo rs
CM CM TH rH
o o o o
CM o in -o
T in n CM
o- rs «r n
o TH CM r*)
0 O O O
1 1 1 1
in rs co in
o o o o
in o in o
rH -H CM
in
o o o o •
o o o o
0000
1
CM CM CM CM
O O O O
o o o o
1 1 1 1
rs N is rs
O 0- 0- O-
rH O O O
-o r*i r*i T
o o o o
1 1 1 1
oo rs CM o
0 CD 0 CM
O O rH rH
O rH CM T
-o rH M rs
CM in o- -o
r*i o 03 -o
O T rs rH
rH rH rH CM
CM in 03 rH
o in o -c
T in rs oo
rs m in in
rH M -O rH
1 1 1 1
1
CM CD 0- 0-
TH CM T -0
CM M T in
1 1 1 1
-0 M rH CM
O rH CM l*>
O O O O
M in CM O
in r*) o T
o o o o
1 1 1
•0 T CM rH
r*i r*) T •HS
o o o o
T T rv (N
O- in rH CM
O- O rH CM
-o in CM in
TH M -0 0-
O 0 O 0
1 1 1 1
rH 00 rH rH
rH M T rH
O O O O
in o in o
rH rH CM
-0
o o o o
o o o o
o o o o
i
CM CJ CM CM
o o o o
o o o o
1 1 1 1
-o -o N rv
CM rH O 0-
o in o- o
IS IS IS 03
o o o o
1 1 1 1
-o in -o n
•0 CM 0 rH
o o o o
I 1
O rH rH CM
O 00 O CM
CM n -o o
rH
O IS M CM
rs 03 O rH
rH rH
O 03 M O
1*1 03 T 03
CM r 4 M r*i
1*1 -0 1*1 rH
TH CM T rs
1 1 1 1
0- M 0- 0-
TH -0 O T
rH rH CM CM
1 1 1 1
T o- in 1*1
O O rH CM
o o o o
T ro TH TH
o o o o
1 1 1
03 -0 03 T
o o o o
O- •« -0 CO
rs rs CD CD
rH I*) O TH
rH CM T -0
o o o o
1 1 1 1
rH rs rH 1*1
o TH r*i 1*1
0 O O O
in o in o
TH rH CM
rs
o o o o
o o o o
0000
rH rH rH rH
o o o o
0 0 0 O
1 1 1 1
M T T T
rs -o -o -o
0 O 0 0
CM 1*1 T T
in in in in
0 O O 0
1 1 1 1
rs CM n is
in in in -o
n r*i r*i M
i i i i
r*i in -o oo
o o o o
r*i o- T o
TH rH CM r*!
n -a oo o-
r*i r*i 1*1 1*1
TH CM o in
CM M T T
rH rH rH rH
o- TOO n
O TH rH CM
1 1 1 1
T n o in
IS 00 O- 0-
i i i i
CM T -0 00
O OO O
o o o o
O CM CM 0-
CM rH O O
O O O O
1 1 1
•o -o in in
CM CM CM CM
O O Q O
rH rH 0- 00
N rs m T
M r*i t*i m
rs pi o rs
O rH CM CM
o o o o
1 1 1 1
O T T 0-
.H O O O
O O O O
1 1
in o in o
rH rH CM
03
O O O O
o o o o
o o o o
1 1
r*i 1*1 ro ho
0000
o o o o
rs is rs rs
rH O O rH
o o o o
oo o- o- w
CM CM CM I*)
1 1 1 1
TH oo CM rs
m o «r in
in in in in
i i i i
rH hO CM O
TH CM T in
rH CM 03 -O
hO -O rH T
rH rH
T 03 O 0-
n co n o
rH rH CM M
TH rv in T
in o CM T
ro in -0 oo
is m T T
CM in o CM
1 1 rH TH
i i
o- is CM in
-o TH n T
CM T in rv
i i i i
-0 O- 1*1 T
O O rH CM
o o o o
rH rs in N
hO rH O T
o o o o
i i
-0 0- 03 rH
rs CM CM in
O rH CM M
O CM 00 T
0- -o T M
~o rs o- CM
rH
o in CM T
CM n in GO
o o o o
1 1 1 1
CO T rs T
rn N in CM
o o CM in
i i i i
in o in o
rH rH CM
0-
265
-------
o
o
o
Csl
I
CQ
1 OO O O
_J 1
O 1
to i -o«r » CM
1 1 1 1 1
t
1 O-IS rs CM
O 1 -0 -0 -0 0*
I «•«• *r i
_l 1 isn CM 0-
UJ A 1 *H ro -0 i-4
en i i i i -.
M 1 1
a i
i oo CM ro o
A 1 -0 W CO 0-
1 ,-4 CM CM CM
1 1 1 1 1
1
1
1
t -0 *-4 rs IS.
in i ox H CM
i oo o o
i
i
1 CM O- CM ^T
L. * 1 CMO •< m -<
M 1 O O O rt
i
Z 1
i o- n in -4
o CM i m co oo o
:> i»»vin
_j t
a i
en i «r t>« » o
to ^ i »H CM *• rs
Q 1 O O O O
i i i i i
i
1 in CM -0 0-
o i o o ~ \ei
1 O O O O
ii ii
Z 1
o 3 i mono
1-4 « 1 -1 •« CM
O 1
U 1 O
K 1 **
O OO O
0 OO O
o o o o
r4 «H CM (M
o o o o
o o o o
in in -o -a
-o - |