WATER POLLUTION CONTROL RESEARCH SERIES
13020 OPB 08/71
 MANAGEMENT OF NUTRIENTS ON AGRICULTURAL
      LAND FOR IMPROVED WATER QUALITY
;NVIRONMENTAL PROTECTION AGENCY*RESEARCH AND MONITORING

-------
        WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters.  They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Environmental
Protection Agency, through inhouse research and grants
and contracts with Federal, State, and local agencies,
research institutions, and industrial organizations.

Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications
Branch, Research Information Division, Research and
Monitoring, Environmental Protection Agency, Washington,
D. C. 20460.

-------
       MANAGEMENT OF NUTRIENTS ON  AGRICULTURAL  LAND

                 FOR IMPROVED WATER QUALITY
                              by
                   Department of Agronomy
          College of Agriculture and  Life Sciences
                     Cornell University
                   Ithaca, New York 14850
                            for  the


             OFFICE OF RESEARCH AND MONITORING

               ENVIRONMENTAL PROTECTION AGENCY
                     Project No.  13020 DPB
                         August,  1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25

-------
                  EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication.  Approval does not
signify that the contents necessarily reflect the vievs and
policies of the Environmental Protection Agency nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
                          ii

-------
                             ABSTRACT

A rainfall simulator was utilized to determine the effects of 2, 10,
and 20 year storm frequencies on losses of water, soil and nutrients
from plots subjected to different crop rotations, fertilizer schemes
and manure applications.  Crop rotations, rates of fertilizer and
manure were compared.  Simulations were made on freshly tilled soil.

Comparative erosion losses were as follows:  continuous sod < corn -
alfalfa rotations < continuous corn.  Fertilizer alone tended to
increase runoff, but this effect was overcome when fertilizer was used
with manure.

Continuous recording of surface and subsurface flow and subsequent
losses of nutrients to the environment was conducted on larger plots.
Rate and time of fertilization determined the plant nutrients lost.
Returning crop residues to the soil improved water infiltration,
increasing deep seepage losses.  Proper timing of fertilizer appli-
cations could control adverse environmental effects.

Phosphorus inputs into cultural media as it related to algal growth was
studied.  Sustained concentration of phosphorus at or above a critical
concentration determined the biomass phosphorus.

This report was submitted in fulfillment of Grant No. 13020 DPB,
Agricultural Pollution Control Unit, Technology Division, Office of
Research and Monitoring, Environmental Protection Agency.
                               iii

-------
                             CONTENTS
Section




     I     Conclusions




    II     Recommendations




   III     General Introduction




    IV     Small Scale Runoff Study




     V     Field Scale Water Quality Study




    VI     Algal Nutrient Study




   VII     Economic Considerations




  VIII     Acknowledgments




    IX     References




     X     Publications




    XI     Glossary




   XII     Appendices
Page




   1




   3




   5




   9




  43




  67




  79




  81




  83




  87




  89




  91
                                 v

-------
                              FIGURES
                                                                Page
 1       Comparison of  Soil Loss  and Runoff  for Various
        Treatments for Continuous  Corn from Bare Soil             15

 2       Effect  of  Fertilizer and Manure on  Runoff for Continuous
        Corn Per  Storm Frequency from Bare  Soil                   16

 3       Loss of Nitrate Nitrogen in Surface Runoff for
        Continuous Corn Per Storm  Frequency from Bare Soil        17

 4       Loss of Soluble Orthophosphate in Runoff Water for
        Continuous Corn Per Storm  Frequency from Bare Soil        18

 5       Effect  of  Fertilizer and Manure on  Soil Loss for
        Continuous Corn Per Storm  Frequency from Bare Soil        19

 6       Loss of Organic Matter in  Sediment  for Continuous  Corn
        Per Storm Frequency from Bare Soil                         20

 7       Loss of Total  Nitrogen in  Sediment  for Continuous  Corn
        Per Storm Frequency from Bare Soil                         21

 8       Loss of Total  Phosphorus in Sediment for Continuous  Corn
        Per Storm Frequency from Bare Soil                         22

 9a     A Ranking by Soil Aggregate Stability of Ten Continuous
        Corn Management Systems  With and Without Manure           23

 9b     The Influence  of Aggregate Stability on Surface Runoff
        and Soil  Loss  from Bare  Soil                              24

10       Runoff  Coefficients Under  Four Selected Rotations  Per
        Storm Frequency from Bare  Soil                            25
             i                                                   *
11       Nitrate Nitrogen Loss in Runoff Under Four Selected
        Rotations Per  Storm Frequency from  Bare Soil              28

12       Soluble PO,-P  Loss in Runoff Under  Four Selected
        Rotations Per  Storm Frequency from  Bare Soil              29

13       Soil Loss in Runoff Under  Four Selected Rotations  Per
        Storm Frequency from Bare  Soil                            30

14       Organic Matter Loss in Sediment Under Four Selected
        Rotations Per  Storm Frequency from  Bare Soil              31
                                 vi

-------
                                                                Page

15      Total Nitrogen in Sediment Under Four Selected
        Rotations Per Storm Frequency from Bare Soil              32

16      Total Phosphorus in Sediment Under Four Selected
        Rotations Per Storm Frequency from Bare Soil              33

17      Percent Water Stable Aggregates in Surface Soil -
        Soil from Four Selected Rotations at Completion of
        Rotation                                                  34

18      Runoff Coefficients for Hay Rotations Per Storm
        Frequency from Bare Soil                                  36

19      Soil Loss for Hay Rotations Per Storm Frequency
        from Bare Soil                                            37

20      Nitrate Nitrogen Loss in Runoff for Hay Rotations
        Per Storm Frequency from Bare Soil                        38

21      Soluble Orthophosphate Losses in Runoff for Hay
        Rotations Per Storm Frequency from Bare Soil              39

22      Loss of Organic Matter in Sediment for Hay Rotations
        Per Storm Frequency from Bare Soil                        40

23      Total Nitrogen Lost in Sediment for Hay Rotations Per
        Storm Frequency from Bare Soil                            41

24      Total Phosphorus Lost in Sediment for Hay Rotations
        Per Storm Frequency from Bare Soil                        42

25      Sequence of Events - 1970-71 - Field Scale Water
        Quality Study                                             46

26      Nutrient and Hydrologic Cycle for Agricultural Land       49

27      Monthly Water Losses to Deep Seepage for Good and
        Poor Management - 4/70 - 3/71                             51

28      Monthly Averages of the Percentage of Rainfall that
        Occurred as Surface Runoff Under Two Management
        Systems - 4/70 - 3/71                                     52

29      Aggregate Stability (%) of Soil on Plots Under
        Different Residue Management and Fertility Levels         53

30      Nitrate-Nitrogen Lost in Runoff Water by Months           55
                                vii

-------
                                                                Page

31      Total Accumulative Surface Runoff Volumes,
        4/70 - 3/71                                               56

32      Total Accumulative Nitrate-Nitrogen, and Soluble
        Orthophosphate Losses from Surface Runoff, 4/70 - 3/71    57

33      Nitrate-Nitrogen Loss to Deep Seepage - Corn -
        4/70 - 3/71                                               58

34      Nitrate-Nitrogen Loss to Deep Seepage - Beans/Wheat -
        4/70 - 3/71                                               59

35      Nitrate-Nitrogen Loss to Deep Seepage - Wheat -
        4/70 - 3/71                                               60

36      Apparent Relationship Between Nitrogen Inputs and
        Losses to the Environment (Surface Runoff and Deep
        Seepage), 4/70 - 3/71                                     61

37      Relationship Between Total Nitrogen Inputs and Nitrate-
        Nitrogen Concentrations in Subsurface Flow and Surface
        Runoff, 4/70 - 3/71                                       62

38      Percentage of Phosphorus Immobilized by the Soil,
        4/70 - 3/71                                               65

39      Effect of Various Maintained Phosphate Concentrations
        on the Growth of Chlbrella pyrenoidosa                    71

40      The Effect of Eliminating Phosphorus Supply on the
        Solution Phosphate Concentration and Growth of Cultures
        of Chlorella pyrenoidosa Previously Maintained at
        Approximately 0.1 pM Phosphate                            73

41      The Effect of Eliminating Phosphorus Supply on the
        Solution Concentration and Growth of Cultures of
        Chlorella pyrenoidosa Previously Maintained at
        Approximately 1.5 yM Phosphate                            74

42      Relation Between Phosphate Addition and Maximum
        Expected Aglal Cell Volume                                76
                               Vlll

-------
                              TABLES


No.                                                             Page

 1        Rainfall Simulation Parameters                          11

 2        Management Treatments for Continuous Corn               14

 3        Management Treatments for Mixed Rotations               26

 4        Management Treatments for Continuous Hay                35

 5        The Effect of Various Maintained Phosphate              70
          Concentrations on the Initial and Final Growth
          Rates of Chlorella pyrenoidosa

 6        Cost and Return Per Ton of Manure Spread on             79
          Continuous Corn Silage Under Good Management
                                 IX

-------
                           SECTION I

                          CONCLUSIONS


The overall study was made up of three related studies.  This integrated
approach has permitted an objective evaluation of the parameters con-
trolling the movement of sediment and nutrients from the land surface
to the water.

Rainfall simulation studies on freshly tilled soil showed that contin-
uous corn grown with mineral fertilizer but without manure results in
a deterioration of soil physical conditions.  This in turn results in
greatly increased soil loss and surface runoff.

With manure, fertilizer and crop residues, continuous corn production
results in a soil physical condition that is no poorer than that
resulting from more conventional corn-small grain-alfalfa crop rota-
tions .  The physical conditions of the soil under corn-small grain-
alfalfa crop rotations tends to improve in direct proportion to the
number of consecutive years alfalfa is in the rotation.  Thus, long
term hay rotations produce the most stable soil physical conditions,
which result in the lowest losses of water, soil and plant nutrients
from the land surface.

A large scale field study has shown that the return of crop residues
and their incorporation into the soil increases infiltration of water
into the soil and reduces surface losses of water and erosional loss of
plant nutrients to the environment.  The correspondingly higher sub-
surface loss of water can lead to nitrate nitrogen leaching, which is
also undesirable environmentally.  By avoiding applications of nitrogen
fertilizer in months when crop uptake is low and deep seepage of water
is high (October - May), and by restricting the quantity applied to
just meet crop requirements, leaching losses of nitrate can be kept to
a tolerable level.  Phosphate transport in subsurface flow was ex-
tremely small.

Transport of phosphorus as a component of  surface  water is far greater
than in subsurface water.  In all instances, total removal of phosphorus
from the system is extremely small as compared to nitrogen.

Algal nutrient studies have shown that the concentration of phosphate
in the growth medium and the total amount of phosphate supplied is
important in determining the extent of algal growth.  A maintained
concentration oflyM phosphate produces maximum growth of Chlorella
pyrenoidosa and a phosphate concentration maintained at 0.1 yM supports
a growth rate which is over 80 percent of the maximum.  It is calculat-
ed  from laboratory studies that to produce a nuisance population of
this algae (10' iP/ml) with an initial phosphate concentration of 0.1
yM, the phosphate must be replenished at least 15 times during the

-------
growing season.  Extrapolation of this information to the natural situ-
ation, it can be concluded that the concentration of phosphate in a
lake may not always be a reliable index of the ability of that lake to
produce nuisance blooms of algae.  If high concentrations pf all
nutrients including phosphate are present, algal production will un-
doubtedly be great.  However, the reverse is not necessarily true.
While a low concentration of phosphate will slow the algal growth rate,
total algal production will depend largely on the degree to which
phosphate is replenished from various sources throughout the growing
season as well as the rate at which phosphate is lost from the trophic
zone of the lake.

-------
                            SECTION II

                          RECOMMENDATIONS

The results of this and other research demonstrate that:  (1) farmers
must apply fertilizer nitrogen directly to the crop.  They should
limit the amount applied to only that which will be taken up by the
crop to which the fertilizer is applied;  (2) phosphorus losses from
the soil are proportional to surface runoff.  They are not directly
proportional to the amount of phosphorus fertilizer applied.  New York
State Extension teaching programs (farmer-educational efforts) are now
utilizing these important principles.

Manure has been demonstrated to have a very beneficial effect on re-
ducing surface runoff.  This property is particularly important with
respect to limiting phosphorus losses.  Methods of utilizing manure
as a source of crop nutrients and as a soil physical conditioning
agent need to be vigorously pursued.  Manure may be a good substitute
for considerable amounts of fertilizer nitrogen and phosphorus.
Particularly important are questions relating to the possible nutrient
losses associated with different handling and application schemes.
Careful monitoring of subsurface seepage and surface runoff are
necessary parts of such an experimental study.

Current trends in the agricultural industry, coupled with growing
concern for environmental pollution, make it imperative that data be
collected and analysed with regard to proper disposal of animal wastes.

On-going research in the area of animal wastes has dealt extensively
with means of altering manures (i.e., dehydration, oxidation, etc.) to
render them more manageable.  It is apparent that the land is the
ultimate disposal site regardless of the type of primary treatment that
the animal wastes may have received.  In order to achieve a cyclic
system of returning nutrients in manures from animals to crops to
animals again with minimum losses, it is necessary to further study the
soil, crop and manure interactions.

It is recommended that the extensive installation used in the field
scale water quality study, described in this report be used to determine
the effect of various manure utilization and disposal practices on
surface and subsurface water pollution.

Animal wastes both untreated and modified by primary treatment should
be evaluated for crop production and land disposal.  The environmental
consequences of currently  evolving disposal practices should be
thoroughly studied and recommendations formulated.

-------
                            SECTION III

                       GENERAL INTRODUCTION

Since the mid-19th century, sanitary engineers have been successfully
operating water treatment plants and waste disposal plants to protect
the health of urban and suburban populations.  The raw water for many
water systems is drawn from streams.  For the last 30 to 40 years,
sediment from agricultural land has been recognized as a major cause
for the deterioration of stream water quality.  Until recently there
has been little or no consideration of the influence of agricultural
chemicals or agricultural nutrients and animal waste upon water quality.
Within the past decade, the use of chemical nitrogen fertilizer has more
than doubled.  A huge feed lot industry has sprung up in the midwestern
and great plains states.  Dairy and poultry industries of the eastern
United States and elsewhere have had their animaJs concentrated into
large centers of operation with a corresponding concentration of animal
waste.  Thus, it has become increasingly difficult to adequately treat
raw water.

The scope of the present project is to evaluate (1) the influence of
different rates and times of application of commercial fertilizer upon
the quality of water that comes from the land,(2) the influence of
animal manures, under various management systems, upon the loss of
water sediment,and agricultural nutrients, and on (3) the influence of
phosphates in water on the growth of algae.

Before going into the body of the report, a few analogies between soil
and water conservation management of nutrients and the management of
waste disposal plants are described.  One of the basic principles that
is constantly used in a well managed waste disposal plant is the
principle of the mass balance.  That is, the input or loading of a
plant must be such that the treatment process will handle the load.
If an excessive load or periodic excesses of load go into the plant,
and the plant cannot handle such loadings, the effluent will then be
irregular in quality.  In many respects, the hydrological cycle has the
attributes of a waste disposal plant.  Seasonal and climatic fluctu-
ations that take place bring about a fluctuation in the loading capacity
of the soil to the extent that the output in terms of water quality may
be erratic or undesirable.  Often these variations in water quality are
due to load; the hydrologic and climatic inputs are not uniform.  We
will equate the soil to the treatment plant in many respects, or we can
equate the soil to certain aspects of the treatment plant.  We know that
for secondary treatment there are usually trickling filter or activated
sludge plants.  In many respects, the soil profile behaves toward the
fertilizer and animal waste that are used in it as the activated sludge
and the trickling filter units behave toward the input or loadings that
are put into these portions of a plant.  To cite some of the more obvi-
ous limitations of the soil and water conservation nutrient management
system, let us consider the facts of rainfall.

-------
Each climatic zone has its own rainfall characteristics.   Parameters
used to characterize rainfall are:  amount, frequency, intensity and
duration.  One must consider the total range of rainfall and preci-
pitation events in order to plan a successful water quality operation
on the land.  Modifying the rainfall or precipitation is very difficult
if not impossible.  By keeping the facts of rainfall in mind it is
often feasible to design the land use and soil receiving system in such
a way that it will maximize both water quality and output.

More should be said in the way of background about the influence of
land use and land management upon the quality of water.  It has been
traditional that a well-managed forest in the eastern United States
will generally produce high quality water.  Many municipalities main-
tain forested watersheds to supply their municipal needs.  This repre-
sents  an   extreme example of control of the watershed or drainage
basin to insure high quality water.  If is often possible to receive
some economic gain from a well-managed forest.  Frequently, it is
impractical and uneconomical to attempt to establish this type of
vegetation over any large area of the land surface.  Where large areas
of grazing land or idle land are properly managed, the water from such
land is frequently of quite satisfactory quality.  This is not
generally true where over grazing occurs.  Cultivated land can be a
source of sediment and salts, which can deteriorate water quality,
because of poor management practices.  The problem is greatly magnified
by the fact that there are numerous individual owners who practice
various levels of management and who, either through carelessness or
lack of knowledge or out of economic necessity, find it impractical or
inconvenient to use proper methods of fertilization, animal manure
application, and erosion control.  It is this complex area of soil,
water and nutrient conservation and crop production that we are seeking
to develop and clarify so that these same areas of agricultural land
will not only serve to enchance the quality of the water emanating from
them, but hopefully, they may also be used as sinks for nutrients from
waste disposal plants that are currently being developed in many small
towns and cities.

High intensity rains of long duration do not cause muddy water to be
discharged out of densely covered woodland or grassland areas.  The
soil surface is protected by vegetation.  Land that is cultivated to
row crops and less dense growing crops can achieve a measure of the
protection inherent in forest and grassland.  This is achieved by
developing a desirable physical condition.  The individual granules or
soil aggregates, by proper management become resistant to the de-
structive action of water.  With a porous, open, water stable soil
surface, fertilizer salts such as nitrates and phosphates can readily
move into the soil.

These salts, and large quantities of water, are thus prevented from
running off the soil surface into surface water bodies.  However, the
fate of these nutrients salts after they enter the soil is highly

-------
variable.  Phosphorus  is generally held in the soil, and extremely
small quantities leave in subsurface water flow.  However, nitrates are
readily leached and will contaminate subsurface water and ground water
to a degree which is dependent on time of application of fertilizer,
crops grown, management, etc.  This study helps to clarify the effects
of these variables.

The body of this report consists of three main areas of work:

 (1)  The small scale runoff studies

 (2)  The large scale water quality studies, and

 (3)  Finally, the effect of nutrients salts on the growth of
     algae in the surface waters.

-------
                             SECTION IV

                      SMALL SCALE RUNOFF STUDY


The preceeding generalizations serve to point out needs and general
trends in the water quality field.  These generalizations are not
quantitative.  For many years, agriculturalists, soil and water con-
servationists, hydrologists, agronomists, and agriculture and sanitary
engineers, have tried to develop quantitative expressions or relations
that could be utilized to predict the quality of water as it comes from
agricultural land.  Because so many professional attitudes and concepts
have been involved, the resulting predictive expressions are varied.

Historically, the first concern was with erosion as a destroyer of
farmland.  Early agriculturalists (Hall, 1937) sometimes gave little
thought to the down stream influences of erosion in terms of sediment
and undesirable nutrients for water quality.  The major concern was on
site loss to the individual farm.  After World War I, a major source
of information developed under the leadership of H. H. Bennett.  Ero-
sion Experiment Stations were developed by Bennett at some 35 or more
locations over the United States.  Many soils were studied (Middleton,
Slater and Byers, 1932).  These "standard stations" made similar mea-
surements on soil and water loss.  Thus, cross comparisons became
possible between various climatic areas and farming systems.  In
addition to "standard comparisons," such as forest versus cropland,
fallow versus cropland, degree and length of slope, contour versus up
and down slope, a very considerable number of watersheds or small
drainage basins were also studied.  These drainage basins or watersheds
ranged in size from a fraction of an acre to several square miles.  In
many circumstances, improved and prevailing practices were studied on
paired drainage basins (Anonymous, 1960).

The runoff plot studies led to the establishment of a Runoff and Soil-
Loss Data Center at Purdue University (Wischmeier, 1955).  Unfortu-
nately, most of these historic studies covered only soil losses
(sediment) and water losses (runoff).  Little or no consideration was.
given to nutrient losses.  Cash grain farming served as a model and was
neglected.  Consequently, management of manure as it might affect the
quantity of sediment and the quality of the water was not considered.

Within the last ten years, an approach to agricultural runoff has
developed that attempts to combine the small plot and/or drainage basin
with the analysis of data collected over larger areas (Amerman and
McGuinness, 1967).  In effect, the assumption is made that the small
plot or drainage basin is a "unit-source watershed."  This concept has
assumed that the larger basin is made up of representative small units.
Therefore, the expense of intermediate or larger basin studies can be
spared by simply combining the representative small plot studies
(Allis, 1962).

-------
This concept of a "unit-source watershed" was applied to an experimental
area that had been developed and utilized by Dr. R. B. Musgrave for a
study of rotations that might be adapted to dairy farming and feed
production in the period 1954 to 1968.  Each rotation carried with it
a definite level of mineral fertilization.  Some rotations were dupli-
cated with and without manure and it was possible to evaluate the
economic return from manure in this type of a system.  These data have
already been published (McEachron, et. al., 1969).

The more complex question is:  Does the cropping program, manuring and
fertilization affect the quality of water and sediment that comes off
the land?  These more complex questions were approached by analyzing
components of the general problem of sediment production and nutrient
loss.  Wischmeier and Smith (1965) summarized information on the so-
called "soil loss equation."  Without dwelling on the details of this
equation, which are available and discussed in the above reference, it
will be said here that Stage 1 of the so-called "cropping management"
factor (Wischmeier 1960)  was studied and this component is usually the
major source of sediment  and runoff on crop land.   By minimizing or
almost entirely eliminating such complex factors as length and steepness
of slope, and by keeping  the study confined to two soil types, it has
been possible to arrive at an evaluation of the influence of past
management upon the soil  itself.
                             10

-------
                        Materials And Methods
  • '. '.i   '.                '•   .  .    .        	1—

A rainfall simulator (Swanson, 1965) was utilized to study runoff and
erosion losses from a range of soil management and cropping conditions.
The artificial rainfall parameters were treated according to the
standards'of Wischmeier (1959).  The rainfall factor, "R", for one half
hour was 25 units.  This value is a product of the total artificial
rainfall energy expressed in foot tons per acre inch multiplied by the
maximum 30 minute rain intensity expressed in inches per hour.

Rainfall simulation plots were selected at random after having elim-
inated plots containing sand pockets and other abnormalities.  These
selected plots were then treated as if a conventional seed bed were
being prepared for corn.  Prior to the normal date of planting, and
after a one inch rain, the plots were covered with a four mill sheet of
plastic.  This plastic was left in place until the rainfall simulation
study was carried out.  At'this time, the plastic was removed.  All
studies were completed by mid-August.

In the actual operation of the rainfall simulator, three one-half hour
rain periods were used consecutively.  Each period delivered 'one and
one-quarter inches of water.  The rainfall simulation parameters are
summarized in Table 1.

Table 1.  Rainfall Simulation Parameters.
  Storm
   R        Amount                            Probability
Rainfall   of Rain    Duration   Frequency   any one year
 Factor    (Inches)    (hours)    in years        %
A1
A + B
A + B + C
25
50
75
1.25
2.50
3.75
0.5
1.0
1.5
2
10
20
50
10
5
        A is equivalent of a 2 year storm, A + B is equivalent of
        a 10 year storm, and A + B + C is equivalent of a 20 year
        storm.
The probability values in Table 1 are based on any one storm.  If the
total rainfall factor R were summed over all rains in central New York,
the summed value of R would equal or exceed 75 every other year.

Determinations were made on the original plot soils for organic matter,
total nitrogen, total phosphorus, orthophosphate, clay, silt, sand,
aggregate stability, and bulk density.  These determinations, except
                               11

-------
for bulk density and aggregate stability,  were again made on the sedi-
ment obtained from the plots.  Total solids were routinely determined
in the runoff obtained from the rainfall simulation.  The significance
of these observations will be discussed in the next section.  All data
are summarized in the Appendix.  Three main types of management were
studied.  They were continuous corn, mixed crop rotations, and con-
tinuous hay.  Aggregate stability tests were also made on all plots in
the rotation study.  A portion of these results are presented in
Figure 9.  Not all plots were used for rainfall simulation.
                              12

-------
                       Results And Discussion

                           Continuous Corn


The seven management treatments studied under rainfall simulation for
continuous corn are shown in Table 2.  Figure 1 illustrates the rather
close relationship between soil loss and runoff.  Total soil loss was
proportional to runoff regardless of fertilizer and manure variables
(Treatments 4, 5, 6, and 7 were combined).  However, as will be
illustrated in subsequent figures, runoff varied among treatments.

Management Treatment 3 utilized 270 pounds of fertilizer and no manure.
Corn residue from grain harvest was returned to the soil.  Figures 2
and 5 illustrate that at all levels of rainfall simulation, heavy
fertilizer application alone, without manure, resulted in as much as
100% increase in soil loss and water runoff relative to plots which
received manure.  The condition of no fertilizer and manure appli-
cations was not included in the original field study planned by Dr.
R. B. Musgrave.

Figure 2 illustrates that runoff was equal or less under manure only
than under either of the other two management treatments (fertilizers
only and fertilizer plus manure).

Figure 3 shows the total output of nitrogen as nitrates in the runoff
during the sequence of three storms.  Nitrates are soluble and mobile
in the soil profile.  It is apparent that nitrate-N was lowest in the
fertilizer plus manure treatment during the initial stages of water
application.  As the soil became more nearly saturated, nitrates from
these soil management treatments moved into the runoff water.  It
suggests that nitrates very likely move through the topsoil layer of
ten inches and then "overflow" into the plot runoff.

Figure 4 shows only very minor differences in PO.-P output.  Perhaps
the soil management treatment of fertilizer only is the highest while
fertilizer plus manure is lower.

Figures 6, 7, and 8 represent the loss of organic matter, nitrogen, and
total phosphorus under the three management system of manure only,
fertilizer only, and manure plus fertilizer.  It is worthy of note that
losses of the constituents from the manure treated plots and the manure
plus fertilizer plots are very similar.  The losses from the fertilizer
only plots are fully 20% higher in organic matter, nitrogen and phospho-
rus than losses from manured plots.

                       Soil Physical Condition

Interest necessarily centers on those materials which come from the
soil and contribute to water pollution.  Yet soil physical condition
plays a major role in controlling rates of infiltration and water run-
                               13

-------
Table  2.   Management Treatments for Continuous Corn
- Treatment
No. Code
I
2
3
4
5
6
7
(Ay)
(Bw)
(Ccy)
(Aw)
(Ax)
(Az)
(Ccx)
Use
Grain
Grain
Grain
Grain
Silage
Grain
Grain
Manure
- (T/A)
6
0
0
6
6
6
6
Fertilizer Total Fertilizer
(Ibs./A) (Ibs./A)
0- 0- 0
30-30-30
30-90-60
30-30-30
30-30-30
30-30-30
30-90-60
(30)1
(90)
(30)
(90)
(90)
(90)
0
120
270
120
180
180
270
   (   )  = Anhydrous Ammonia Sidedress

-------
     S-i
     O
     
-------
          o
          (3
          C
          
-------
                 3.0
                 2.5
              Qj
              o  2.0
                 1.5
                 1.0
                   .5
•   Manure only
O   Fertilizer only

    Fert. plus manure
                     0
           2 year
            storm
10 year
 storm
                                                                20 year
                                                                 storm
Figure 3.  Loss of Nitrate Nitrogen in Surface Runoff for Continuous
                    Corn per Storm Frequency from Bare Soil
off and soil loss from treatment number three was approximately 60%
more than number seven which had received six tons of manure (see
Figure 1).  Figure 9b illustrates the influence of aggregate stability
on runoff and soil loss for the two year storm frequency and contains
a large amount of variation.  All manure treated management systems for
continuous corn out rank continuous corn without manure in terms of
water stable soil aggregates.  Management systems 4 compared to system
7 suggests that manure additions increase water stable soil aggregates
by at least 25% at high levels of fertilization.  At moderate levels of
fertilization there is lesser difference in this aspect of soil physical
condition.  Contrast management systems 5 and 6 which both receive
identical amounts of fertilizer.  Reference will be made to soil
management as it concerns soil physical condition under economic con-
siderations.
                               17

-------
           o
           o
           PM
            -
           O
           CM
3.0


2.5


2.0


1.5


1.0


 .5
                      •  Manure only
                      O  Fertilizer only
                      O  Fert. plus manure
                                 2 year
                                  storm
                               10 year
                                storm
20 year
 storm
Figure 4.  Loss of Soluble Orthophosphate in Runoff Water for Continu-
                  ous Corn per Storm Frequency from Bare Soil


                        Mixed Crop Rotations

The management situations and crop sequences available for this study
are given in Table 3.  Four rotation management systems listed in
Figure 10 were selected as being representative of this group.

The percent total suspended solids listed on Figure 10 suggests that
these four rotations are all very similar in the amount of suspended
material per unit of surface runoff from the small test plots.

Two year storm runoff coefficients for these four management systems
average 4%.  The number 4 system with no commercial fertilizer and
only manure gives the lowest runoff at this initial stage.  It is
interesting to follow through to the 10 year storm frequency and again
compare the four management systems.  Here it is clear that number 1
mixed crop rotation containing corn and oats with two years of alfalfa
                              18

-------
        20 year  storm
                        100         200           300
                             Total Fertilizer (Lbs./Acre)

Figure 5.  Effect of Fertilizer and Manure on Soil Loss  for  Continuous

                   Corn per  Storm Frequency  from  Bare  Soil
                               19

-------
          0)
          4-1
          O
          •H
             100.0
              80.0
               60.0
               40.0
          £f    20.0
 •  Manure only

 O  Fertilizer only

 ^  Fert. plus manure
0
2 year
 storm
10 year
 storm
                                                              20 year
                                                               storm
Figure 6.  Loss of Organic Matter in Sediment for Continuous Corn per

                      Storm Frequency from Bare Soil


and modest applications of fertilizer is some 4% to 5% lower in runoff
than the other three systems.  It is suggested that this may be true
because system number 2 has no manure and 50% more commercial fertili-
zer.  It has also received 90 pounds of nitrogen as an anhydrous
ammonia sidedress.  There are indications from the literature that
nitrogen fertilization may decrease infiltration rates.

Management systems numbers 3 and 4 each contain two years of alfalfa as
do systems numbers 1 and 2.  But these two years of alfalfa are
separated by a plowing for wheat.  Thus in numbers 3 and 4, based on a
five year crop rotation, there were three more plowings since 1955 as
compared with numbers 1 and 2.  Also, they have shorter periods of
alfalfa  (one year) on the land.

However, the 20 year storm frequency results show a marked contrast.
Here again management system number 1 shows superiority over the other
systems.  Its runoff coefficients is 25% as compared to number 4 at 37%
while management systems numbers 2 and 3 show runoff coefficient
between these two extremes.  An explanation is suggested for the
                               20

-------
              0)
              M
              CD
6.0


5.0


4.0
  •  Manure only
~ O  Fertilizer only
  \>  Fert. plus manure
              a   3'°
              §
              bo

              I   2-0
              O
              H
                  1.0
                                   2 year
                                    storm
                                10 year
                                 storm
                                          20 year
                                           storm
 Figure 7.   Loss of Total Nitrogen in Sediment  for  Continuous  Corn  per
                       Storm Frequency from Bare Soil


 increased  runoff  from  management  system number  4.  As Table 3 indicates
 this management system receives no commercial fertilizer.  Because
 manure  is  low  in  phosphorus  this  system had produced low yields of
 alfalfa and hence return of  organic matter  as root residues was limited
 relative to the other  management  systems.   This  in turn led to a weaker
 aggregate  stability  (Wischmeier,  1960b).
       \
 The NOo-N  losses  under the four various management systems are presented
 in Figure  11.   The two year  storm output is remarkably similar for each
 of the  four systems.   Apparently  very similar factors are operating.
 Small quantities  of water are moving low concentrations of nitrate.
 During  the 10 year storm NO-j-N output is, apparently, inversely related
 to soil physical  conditions.  The percent of water stable aggregates in
management system number 1 is 74.5.  That for number 4 is 78.3 while
 systems numbers 2 and  3  are  62.9% and 65.2% respectively.  It can be
postulated that with better  aggregation, water and nitrates move down-
ward more freely.  Hence there is a lower nitrate loss in surface
                                21

-------
                2.0 -
             to
             CD
             O
             js
             CM
             o
             H
Manure only
Fertilizer only
Fert. plus manure
                                 2 year
                                  storm
                      10 year
                       storm
20 year
 storm
Figure 8.  Loss of Total Phosphorus in Sediment for Continuous Corn per
                          Storm Frequency from Bare Soil
runoff.  In the case of the 20 year storm, enough water has been applied
to cause movement through the topsoil or plow layer.  Possibly in
management system number 1, two, years of alfalfa plus .manure reduced
runoff and increased infiltration to the point where even the 20 year
storm will not cause appreciable quantities of water to move laterally
through the topsoil.  Management system number 2 without manure has a
lower nitrate concentration than systems numbers 3 and 4 with manure.
The respective concentrations expressed in parts per million in water
coming off are: 5.2 versus 8.2 and 11.4.  The differences in concen-
tration together with the runoff differences in addition to the ease
of movement of water through the plow layer (great in the ease of
management systems numbers 3 and 4) account for the differences in
NOo-N output per acre.

The case for PO^-P (Figure 12) can be explained more simply.  Under the
2 year rain there was little or no PO/-P delivered.  In the case of the
                               22

-------
                CD
                0)
                4J
                cfl
                00
                cu
                )-(
                M
                60
                O

               , Cl)
               .-•i-H

                CO
60-


50


40


30
                "£  ,20
                    10
No Manure
** -.
-
_




—
"





x-s
o

\^ _J

o
^D
|
£^
O"N
1
0
CO





••-•'








0
CO
'• .f
' O
v^
- 1
(^5
ON
1
o





"• f

."*,






o

rH

p
vO
•..:i..

o\
I
0
CO













o
o>

o
vO
f
O

1
0
CO






jv





-'
o
CO

o
CO
A i
o
• CO
1
0
CO
6 Tons Manure

•














o
CO

o
CO
1
O
CO
1
o
CO
X




_' ;










0
o\

0
\^
1
o
o\
1
o
CO















0
m
rH

0
CO
1
O
CO
1
o
CO


















o

1

o

1

o
















0
a\

o
CO
1
o
CO
1
o
CO
                                                                    10
                                      Management System
Figure  9a.   A Ranking by Soil Aggregate Stability of Ten Continuous

                     Corn Management Systems With and Without Manure.
  • ^-,  .-,       ^        ^ .. .     -   ;  . •   ..    . .   '   •     .   „       -  .


10 and  20  year storm simulations the PO/-P delivered under both storm
intensities  was somewhat proportional to the annual phosphorus and
organic -mattef inputs.  These were by systems of 'management : #1 - 30
pounds  plus  manure; '#2 - 45 pounds, -no manure; #3 - 30 pounds plus
manure;  and ' #4 - no  phosphorus plus manure.  If a'value of 2.5 pounds
of phosphorus, expressed as
          ov
                                   is assigned to each ( ton of the six
.ton manure  application,  then each manure application would carry 15
pounds  of P^Oc with it.  ! Because mineral fertilizer phosphorus is fixed
on the  soil clay and organic phosphorus from manure is not, it would
undoubtedly be safe to assume greater effectiveness of the organic
phosphorus  in supplying  PO^-P.
                                23

-------
         o
         o
         o
            40
            30
14-t
M-l
°  20
         (U
         O
         M
         CO
Figure 9b.
            10
                    16.5
                 Runoff
                      30          40          50

                            Percent Aggregate Stability
                                              60
                                                                  4 -
                                                      3 -
                                                                  2 -
                                                                  1 -
                                                                      o
                                                                      o
                                                                       c
                                                                       o
                                                                      H
                                                           w
                                                           CO
                                                           0
                                                          (J
                                                          •rl
                                                          O
                                                          CO
The Influence of Aggregate Stability on Surface Runoff and
               Soil Loss from Bare Soil
                              Soil Loss

Under 2-year simulated rain, soil losses are insignificant (Figure 13).
In the case of 10 year and 20 year simulated rain, the soil losses
expressed in tons/acre are proportional to the runoff coefficient.

Organic matter losses (Figure 14) are easily understood if the concen-
tration of organic matter in the runoff suspension is considered.
Under the two year simulated storm conditions management system number 1
contains 6.6% organic matter in the runoff suspension.  Management
system number 2 contained only 3.9% organic matter.  The manure applied
to management system number 1, apparently, greatly increased the  con-
centration of organic matter in the runoff.  The fertilizer used  in
management system number 2 actually decreased the organic matter
                               24

-------
M-l
O

1
8
o
•H
<4-l
14-1
(1)
O
  M-l
  14-1
  O
  c
     40
     30
     20
     10
          1.   C-O-A-A with manure
          2.   C-O-A-A no manure
          3.   OO-A-W-A with manure
          4.   C-0-A-W-A with manure
                                        Fertilizer

                                      83 Ibs.  per year
                                     120 Ibs.  per year
                                      81 Ibs.  per year
                                      00 Ibs.  per year
#1  #2  #3   #4
                             #1   #2  #3  #4
           #1  #2   #3   #4
          nnfln,
            2 year storm
                             10 year storm
20 year storm
Figure 10.  Runoff Coefficients Under Four Selected Rotations per Storm
                           Frequency from Bare Soil

content of  the runoff suspension.  Management system number 3 also
generated less organic matter content in  the soil of the eroding suspen-
sion than in the original  soil (4.3% vs 4.6%).  Hence the lower organic
matter loss under this system.  There was no sample from system number
4 due to inadequate sample size.

The same trends discussed  under the two year storm frequency for
differences in organic matter between the original soil and the sus-
pended soil in runoff hold true for the 10 and 20 year storms.  Manage-
ment system number 1 organic matter content of soil in runoff exceeds
management system number 2 by more than 10% on a comparative basis.
                             25

-------
      Table 3.   Management Treatments for Mixed Rotations
ISJ
Manure
Code Rotation (T/A)
C.W (1) C-O-A-A
Corn
Oats
Alfalfa
Alfalfa*


C4Y (2) C-Q-A-A
Corn
Oats
Alfalfa
Alfalfa*


C,Z (3)' C-O-A-A
Corn
Oats
Alfalfa
Alfalfa


12
0
0
12



0
0
0
0


.-
0
0
0
0

Fertilizer Total Fert.
(Lbs./A) (Lbs./A)
0
30-30-30 (30)
30-30-30
0-30-30
0-30-30
Mean

•.:•.:
30-30-30 (30)
30-30-30 f
0-30-30
0-30-30
Mean


30-90-60 (90)
30-30-30
0-30-30
0-30-30
Mean

120
90
60
60
83


120
"90
60
60
^83
r. -x
270
270
' 90
60
60
120 „
Manure
Code Rotation (T/A)
D-Y (2) C-C-0-A-A
Corn*
Corn
Oats
Alfalfa
Alfalfa

E Y (1) C-0-W-A-A
Corn
Oats
Wheat
Alfalfa
Alfalfa*

F.W (1) .C-0-A-W-A
Corn "
Oats
Alfalfa
Wheat
Alfalfa*

0
0
0
0
0


10
0
10
0
10


10
0
0
10
10
Fertilizer Total Fert.
(Lbs . /A) (Lbs . /A)

30-30-30
30-30-30
30-30-30
0-30-30
0-30-30


30-30-30
30-30-30
15-30-30
0-30-30
0-30-30


30-30-30
30-45-45
0- 0- 0
15-45-45
0-30-30

90
(30) 120
90
60
60
Mean 84

(30) 120
90
(30) 105
60
60
Mean 87

(30) 120
120
0
105
60
                                                                                                Mean  81
                                                                                     Continued on next page

-------
      Table 3.   (Continued)
Code Rotation
Manure Fertilizer Total Fert.
(T/A) (Lbs./A) (Lbs./A)
D W (1) C-C-0-A-A
Corn*
Corn
Oats
Alfalfa
Alfalfa

10
10
0
0
10

30-30-30
30-30-30
30-30-30
0-30-30
0-30-30

90
(30) 120
90
60
60
Mean 84
Code Rotation
Manure
(T/A)
Fertilizer Total Fert.
(Lbs./A) (Lbs./A)
F5Z (2) C-0-A-W-A
Corn
Oats
Alfalfa
Wheat
Alfalfa*

10
0
0
10
10

0- 0- 0
0- 0- 0
0- 0- 0
0- 0- 0
0-0-0.

0
0
0
0
0
Mean 0
NJ
             1. * = Crop plowed under for this experiment.




             2. ( ) = Anhydrous Ammonia Sidedress

-------
     0)
     S-l
        3.0
     M-l
     M-l
     O
     a
     •H
     a
     a>
     60
     o
     M
     4-J
     td
     M
     4-1
2.0
        1.0
          0
                                                     #1   #2   #3  #4
                          #1   #2    #3   #4
              #1  #2   #3   #4
                                 O.
                2 year storm       10 year storm      20 year storm

Figure 11.  Nitrate-N Loss in Runoff Under Four Selected Rotations per

                           Storm Frequency from Bare Soil

This same comparative basis holds when management systems numbers 3 and
4 are compared.

Total nitrogen loss is presented in Figure 15.  These values follow
organic matter losses very closely.  Total nitrogen contents are
approximately 5% of the organic matter content.  Nitrogen is lost in
the erosion processes primarily as soil organic matter.

Figure 16 presents the total phosphorus losses for these four systems.
Again these losses closely parallel the organic matter losses.  Total
phosphorus losses are approximately 2% of the organic matter losses.
                               28

-------
        2.0
  o
  I
  c
  •H
  O
  O
    
 •§
 rH
  O
 CO
1.5
1.0
         .5
  0
                                             #1  #2   #3
                                  #1  #2   #3  #4
              #1  #2   #3  #4
                2 year storm      10 year storm       20 year storm

Figure 12.  Soluble PO^-P Loss in Runoff Under Four Selected Rotations

                       , Per Storm Frequency from Bare Soil

                     Soil 'Physical Condition

Figure 17 shows the percent water stable aggregates for each management
system under the corn crop.  Except for management system number 4
these values are inverse to the runoff coefficients.  Management system
number 4 generates increased runoff only under the 20 year storm
values.  With the 2 year storm the runoff is in good alignment with
aggregate stability.  it must be concluded that the water stable ag-
gregates  generated under system number 4 are less resistant to rain
drop impact than under the other three systems.

                      Continuous Hay Rotations

As a group (Table 4) these management systems are unique.  They show
low runoff coefficients (Figure 18) and low soil losses (Figure 19).
In the group of the three management systems presented here number 3
treatment shows double the runoff over the other two management systems.
This difference can be explained by the'fact that number 3 is
essentially a "neglected meadow" system of management.  Although dry
                                29

-------
        2.0
        1.5
     co
     §  1.0
     o
     1/3
          .5
                                                      #1  #2  #3   #4
                                  #1   #2  #3  #4
              #1   #2  #3   #4
                                   n
          0
               2 year storm
                            10 year storm
20 year storm
Figure 13.  Soil Loss in Runoff Under Four Selected Rotations per Storm
                            Frequency from Bare Soil
matter
systems
rooted
in root
runoff
carries
soil at
production under this system is 90% of the numbers 1 and 2
  the character of vegetation is far different.   It is shallow
and more sparse than alfalfa.  As a result of this difference
 density the number 3 management systems is subject to continued
under continued high rain intensities.  This increased runoff
 both NO -N (Figure 20) and PO^-P (Figure 21) with it out of the
 higher volumes of water (Wischmeier, 1960b)
Soil loss under 10 year storm intensity is much greater with management
system number 3 than numbers 1 or 2.  It would appear that management
system number 3 generates a soil physical condition that resists
erosion up to a certain critical value.  Even though the system has
received manure it then erodes to a considerable degree although less
than continuous corn without manure.  The above is also true for the
20 year simulated storm.

Organic matter losses (Figure 22) parallel soil losses.  Again total
nitrogen (Figure 23) makes up approximately 5% of the organic matter.
Total phosphorus (Figure 24) is approximately 2%.  For the three
                                 30

-------
u
-^.
Jg 100
1-3
V-X
4-1
f"H O/\
g 80
o>
w 60
•H
S 40
4-1
£
| 2°
60
l-l
o

-




-

-
#1
'n







#1 #2' #3

#2 #3 #4
nm_,







MDBIIII^
















#4

—

#1 #2 #3 #4

































































               2 year storm         10 year storm      20 year storm

Figure 14.  Organic Matter Loss in Sediment Under Four Selected
                  Rotations per Storm Frequency from Bare Soil
management systems the percent of water stable aggregates was very high,
having a mean value of 70%.  There was no appreciable difference
between number 2 management system without manure and the other two
systems with manure.

Additional comparisons from the data were made between continuous corn
for silage, continuous corn for grain, corn-corn-oats-alfalfa-alfalfa
and wheat-alfalfa-alfalfa-alfalfa-alfalfa all receiving six tons of
manure per acre per year.  The intent was to determine if the effect of
manure would tend to mask the differences between continuous corn and
rotations consisting of varying amounts of alfalfa.  The results showed
that additions of manure on continuous corn were beneficial to the point
that surface runoff and soil losses were essentially the same as a
mixed rotation (C-C-0-A-A), but that nutrient losses were somewhat lower
for continuous corn.  Long term alfalfa (W-A-A-A-A), as previously
demonstrated, was the most resistant to erosion losses.
                               31

-------
         6.0.
      w  5.0
      c  4.0
      
-------
  4-1
  (3


  I
  -O
  QJ
  to
    s~\
  C 0)
  •H M
    O
  w 
-------
81)
w
at
E 60
CU
60
3 40
en

-------
Table 4.   Management Treatments for Continuous Hay
Code
Rotation Manure
(T/A)
Fertilizer
(Ibs./A)
Total Fertilizer
(Ibs./A)
        (1) W-A-A-A-A
        Wheat            10     15-30-30
        Alfalfa           0      0-30-30
        Alfalfa          10      0-30-30
        Alfalfa*          0      0-30-30
        Alfalfa          10      0-30-30
                                 (30)'
                                                 Mean
                       105
                        60
                        60
                        60
                        60
                                                 69
        (2) W-A-A-A-A
        Wheat             0
        Alfalfa           0
        Alfalfa           0
        Alfalfa*          0
        Alfalfa           0
                        15-30-30 (30)
                         0-30-30
                         0-30-30
                         0-30-30
                         0-30-30
                       105
                        60
                        60
                        60
                        60
                                                 Mean
                                                 69
  3

  JW
Continuous Hay    6
0-45-45
90
               Crop Plowed under for this experiment.

               ( ) = Anhydrous Ammonia Sidedress.
                               35

-------
     (3
     01
     •rt /~N
     O <4-l
     •rl M-l
     "4-1 O
     H-l fi
     
-------
           CO
           CO
           o
           •H
           O
           CO
               .50
               .40
.30
.20
                ,10
                      #1   #2   #3
                      #1   #2   #3
                   0    2 year
                          storm
                       10 year
                        storm
                                      #1   #2  #3
20 year
 storm
Figure 19. \ Soil Loss for Hay Rotations  per  Storm Frequency from Bare

                                  Soil
                                37

-------
2.0
4-4
M-l
§ 1-5
3
(3 O
H i-0
M CO
1 1 ^^
•M td
*|H hJ
t?* ^^^
(U
« -5
4J
•H
a
/\
#1 #2 4


#1 #2

#1 #2 #3 | 	 1
i 1 	 1 1 1
#3
•••^•a




1
MBMBWB





m^^^m







(3






                   \f
                      2 year storm    10 year storm  20 year storm

Figure 20.  Nitrate-Nitrogen Losses in Runoff for Hay Rotations per

                          Storm Frequency from Bare Soil
                              38

-------
         0)
         4J
         (0

         &.O
         M O <-s
         O r-l 01
                2.0 -
1.5
         CU
         o
             U
         M I  0}  1.0
         O

         (U PM
         O
         C/3
 .5
#1 #2
#1 #2 #3
< 	 > -— - .-— - ' ' ' ' J
#3
^^••••m
#1
,n
#2

/;
'3
                                               —_——	
                     2 year storm     10  year storm  20 year storm
Figure 21.   Soluble Orthophosphate Losses  in Runoff for Hay Rotations

                      per Storm Frequency from Bare Soil
                                 39

-------
                en

                iJ
                .u
                C
                (U
               TJ
                
-------
               4-1
               C
               0)

               •H
               "O
               UJ
               CO ^-v
                  QJ
               C  ^
               •H  O
                60 10
                O ^3
                O
                H
                                                         #1   #2   #3
                                          #1   #2  #3
  //I  #2
  —* r~~

2 year storm
                                         10 year storm  20 year storm
Figure 23.   Total Nitrogen Lost  in  Sediment for Hay Rotations  per

                          Storm Frequency from Bare Soil
                               41

-------
             c
             CD
             s
C 0)
•H J-l
  O
co 
-------
                              SECTION V

                   FIELD SCALE WATER QUALITY STUDY


In the general introduction, we considered the similarity between waste
water treatment plants and the natural hydrological cycle as it
occurred outdoors in connection with soil and water and nutrient
management.  Here we attempt to deal with this problem in that we will
describe in some detail the general philosophy and the method that has
been used to approach the problem so that the findings could be used
over relatively large areas.

The outstanding characteristics of the agricultural soil in the
eastern and central United States is that they are gently sloping, that
for the most part they are silt loams, and that they are subject to
moderate degrees of erosion under cultivation.  It has been well
established that in order to study such phenomena as the rate of move-
ment of water through the soil in the process of underdrainage, a
relatively large area is needed to achieve valid results.  It is also
well-known that the quantities of soil that are removed in studies of
natural rainfall and in studies of artificial rainfall applications,
are dependent upon the size of the area used for the study.  Thus, if
we were to have a valid "ecosystem", a unit in which we can measure the
inputs and effects of various factors upon the soil and the environ-
ment, it would be necessary to have a relatively good sized area so
that we could obtain results that would be quite similar to those that
we might find in a farmer's field.

It also became apparent that it would be highly desirable, as a first
approximation in this study, to emphasize differences between what we
will refer to throughout this text as "good" and "poor" practice.  We
established good and poor practice in a situation that would normally
be considered as cash crop farming.  While cash crop farming is, at
the moment, probably not a major contributing source of pollution in
New York State, it has been and may be in the future, a major con-
tributing source of pollution throughout the mid-west.  It was for
this reason, the two years of preliminary results indicated what harm
excessive fertilization might do to water quality; we have used 24
large plots to study primarily the effect of fertilizer, crop, and
management.  These plots are unique in that they have been treated
under good and poor management consistently since 1956.  The soil types
are representative of fairly large areas.  The amount of drainage water
that percolates through these plots are intercepted by the drain is a
good representation of interception type drainage found throughout the
eastern and mid-western part of the United States.

In most other studies of this type, individual random drains have been
utilized and nitrates and/or some particular component such as ammonia,
have been monitored for irregular periods in connection with the flow
                               43

-------
from a particular drain.  This study has made it possible to relate
nutrient output (NOg-N, NH,-N, PO,-P) to specific cropping practices
and specific management practices and specific levels of fertilization.
It serves as a guide for recommended levels of fertilization for
commercial farmers in New York as well as in the mid-western and
eastern states generally.
                            44

-------
                        Methods And Materials


This experimental installation was designed to study the quantity and
quality of surface and subsurface water flow.  The experimental field
is made up of approximately 30 acres of a Lima-Kendaia soil associ-
ation at the Cornell Agronomy Research Farm near Aurora, New York.  A
total of 24 plots of 0.8 acres each were installed in 1956.  The study
was maintained as a drainage study until 1968.  Half of the plots
appear on the Kendaia silt loam, a somewhat poorly drained soil, the
remainder are situated on the Lima silt loam which is moderately well
drained soil.

Surface runoff and subsurface drain flow has been studied under crop
and management systems that were identical with cash grain farming.
All plots are farmed across the slope with a minimum of tillage.  The
three major treatments considered are crop, fertility, and management.

The crop rotation consists of corn, beans, and wheat each under normal
and high fertility levels.  Each treatment is further divided into two
management schemes.  Good management consists of using cover crops
wherever possible and returning all plant residue.  Poor management is
effected by omitting all cover crops and removing all plant residue
after harvesting.  The cropping and fertilizing system is presented in
Figure 25.  The treatment schedule appears in Table 1 of Appendix B.
The statistical model used is a completely randomized design consisting
of a three factor analysis containing two replications.  The three by
two by two factorial analysis is made up of three crops, each at two
levels of fertility, and each at two levels of management, replicated
twice for a total of 24 plots ([3x2x2]x2).

Surface water is controlled by a series of small interceptor and broad
shallow runoff ditches.  Individual plots have surface slopes on an
average of two to four percent.  Runoff water is diverted into a flume
so that total quantity of surface flow may be measured.  An integrated
sample of the surface runoff is collected for laboratory analysis as
follows:  all surface water is diverted from the interceptor ditch
through an intake grate.  The water then flows into a 12 inch H-flume
where the height of the water in the flume is recorded automatically
against time with an FW-1 stage recorder.  A subsample of approximately
1% is collected by a mechanically driven Coshocton wheel.  This sample
is further divided by a splitter arrangement which collects 20% of the
subsample.  This integrated water sample, taken over the entire period
of flow, is collected in a storage tank.  A liquid and solid (soil
sediment) sample is collected and analyzed as soon as possible after a
storm (see Appendix A).  The data points are transferred from the hydro-
graph to computer cards and a program has been established to calculate
flows for given time periods (see Appendix B, Tables 2, 3, and 4).

Subsurface flow of water is studied with the use of a single drain tile
approximately three feet below the surface.  These tile lines are
                               45

-------
                Wheat
Months of Hydrologic Year




       Beans/Wheat
Corn
s 	 ^^<^^
li 	 J.

*
gj
sr

rt

sr
fu
)-!
^
(D
03
It
fD




















^~ ^
M

A.
*T3 £
M sr
O fD
rt CD
03 rt

Hi
fD
i-i
rt
H-
(— »
P-
N
fD
P-

(— •
O
^

Hi
fD
i-i
rt
P-
h"1
P-
rt
**^




^r^
A 	 0

^
gj
sr
fa
(il
rt

•a
Co
s>
rt
fD
a.

pj
h-1
1—

•a
[— •
o
rt
CO

Hi
fD
h"J
ft
H-
h--
P-
N
fD
a
• — -

1

H-
0<3
sr

HI
fD
I-i
rt
H-
M
H-
rt
Vj

•a
i— i
o
rt
CO

Hi
fD
hi
rt
H-
(— *
H-
N
fD
PJ



^
s
1 	
1
a1
fD
§
CO

j^*
fB
^
fD
CO
rt
fD
Pu



















" 	

f
sr
H-


Hi
fD
I-i
rt
H-
h--
H-
rt
*-<^

•X3
t-1
O
rt
CO

Hi
fD
h^
rt
P-
1—
P-
N
(D
a.



— -^
J

	
cr
fD
§
CO

>a
(U
p
rt
fD
P-

pj
I— '
I—

TJ
H
O
rt
CO

Hi
fD
l~j
rt
P-
H>
P-
N
fD
CL
^~\r
M A
i 	 l_
t
sr
p-


HI
fD
H
rt
P-
I-1
P-
rt
Vj

*O
^— A
O
rt
CO

Hi
fD
l-f
rt
P-
1— •
P-
N
fD
P*



^^^••••••« \J • • •
' 1

n
o
3

sr
03
3
fD
CO
rt
(D
CL




















_^-^-
..J
1 ,
i
03
H-
fD
Pu
H
fD
CO
0)

sr

en


HI
fD

rt
P-

H-
rt


T3
h- '
O
rt
CO




— —• •

^
n
o
i-i
3

*o
H1
3
rt
fD
a.

03
!_i
l_

Hi
fD
f-S
rt
H-

P-
N
fD
a.







^^
M A
_i 	 1
1
sr
p-
OQ

Hi
fD
H
rt
l_i«
l«i
P-
rt


13
}_i
O
rt
03

Hi
fD
I-i
rt
P'
I-1
P-
N
fD
CX



Figure 25.  Sequence of Events - 1970-71 - Field Scale Water Quality Study

-------
centered in 12 randomly selected plots.  The tile is four inches in
diameter and empties into an underground metering tank.  The tile flow
empties directly into a buffer chamber before flowing through a 90°
sharp crested V-notch weir.  A Cassela recorder plots the stage height
versus time in the form of a hydrograph.  Tile flow data is also trans-
ferred to computer cards and a program calculates gallonage of tile
flow per plot.  A representative subsample of tile flow is collected
from a uniform drip flow into a container.

Analytical determinations for ammoniaical-nitrogen, nitrate-nitrogen,
soluble orthophosphate and, since January 1971, total phosphorus in
water, are done at the Cornell University soil testing laboratory.
Suspended solids in solution represented as sediment samples, are
collected and weighed  (see Appendix A).
                                47

-------
                       Results And Discussion


Even though a watershed may have a large capacity to store water in the
soil, the infiltration rate may limit the percentage of rainfall that
can enter the soil.  Whenever rainfall occurs more rapidly than infil-
tration, water moves downslope across the surface.  A portion of the
water that does infiltrate may only percolate to a shallow depth and
reappear downslope as ground water seepage at the surface, otherwise
percolating water, if in excess, will leach through the soil profile
to the ground water reservoir, or to a subsurface drain (Figure 26).

Whether water moves across the surface or through the soil profile, it
will transport material.  Surface runoff, unlike subsurface flow, can
transport solids as suspended material.  Transport of solids can be of
considerable magnitude especially during heavy rains when quantities of
soil lost can be very high.  Phosphorus, being relatively easily fixed
by the soil and rather insoluble, can be lost in large amounts by the
physical removal of soil from the land surface.  Nitrogen is readily
converted to a soluble form and is easily subject to losses by both
surface runoff and leaching (Russell 1961).

Any farm management scheme that can promote a reduction in surface run-
off and a consequent increase in infiltration is effectively controlling
water pollution from surface runoff.  This idea had been incorporated
into the water quality study as a main treatment, which is designed as
good and poor management.  In this particular study, good management
consists of returning crop residues back to the land and on poorly
managed fields, the crop residue is removed.  The practicality of this
type of treatment can be illustrated by considering a corn field.  A
poorly managed system can be corn cut for silage in which almost all of
the crop is removed, in contrast to corn harvested for grain where
almost all of the crop except the grain remains in the field.

The return or addition of organic matter to the soil promotes soil
improvement.  Organic matter, whether plant residues or manure, has
been shown to have an effect on improving soil structure (see previous
section).  If soil structure is improved, there is better aggregation
and a consequent increase in infiltration.  This effect is desirable
for surface water pollution, but increases the risk of contaminating
subsurface water with leached nitrates.

The following figures illustrate the main findings of the study.  They
confirm many predictions based on the results of the small scale com-
ponent studies described in the previous section.  The hydrologic year
used in this study is based on Brakensiek's (1959) demonstration that
soil moisture is the most predictable at about April 1.  Deep seepage
values were calculated from Thornthwaite's (1957) extensive work on
soil moisture storage, evapotranspiration and rainfall using actual
measured values for rainfall and surface runoff.  The tile drain acts


                              48

-------
.p-
VD
                                Human

                             Consumption
         Animal

       Consumption
           Top soil
           Sub soil
                                                                                      Fertilizer  [
       Evapotranspiration
       Volatilization,
       Denitrification
                                                                                       Runoff*
                                                       Deep seepage*

     Figure 26.   Nutrient and Hydrologic  Cycle for Agricultural Land
                                                                                          Drain flow*
* Undesirable Loss of
  Nutrients to the Environment

-------
as an indicator of potential deep seepage conditions, and supplies a
sample of percolate for analysis.

                         Hydrologic Effects

It has been shown that deep seepage occurs mostly between October and
December (see Figure 27).  The only other time that deep seepage may
be of concern is in March when the effect of accumulated snow is most
noticeable.  This is highly variable from year to year.  However, it
is almost certain that by June 1, in most years, any appreciable deep
seepage will have ceased.  During the fall of 1970 frequent heavy rains
caused unusually high deep seepage.

The effect of management on deep seepage is the inverse of its effect
on surface runoff.  Figure 28 shows the decrease in runoff which results
from increasing organic matter returns on the well managed plots.
Figure 28 also shows the effects of the two management systems on the
percentage of rainfall that runs off the land.  Data is only presented
through December because snow melt is hard to interpret with regard to
random snow drifting and runoff.  These monthly averages illustrate
the effect of good management in reducing overland flow.  The effect on
deep seepage is clear in Figure 27, but in general, the subsurface flow
is greater than surface runoff and the effect is not so dramatic.

Aggregate stability, the resistance of soil particles to destruction by
water action, is increased by organic matter returns, confirming effects
seen on the small scale study (see Figure 29).  It is this aggregate
stability, together with soil cover on the well managed plots which
decreases runoff on these plots.  The effect of the type of crop grown
on deep seepage flow appears to be negligible.

                              Nitrogen

Nitrogen is present in the soil mainly as the organic form incorporated
with the soil organic matter, as ammonium, and as nitrate.  In the
latter form, it is highly mobile and most readily lost from the soil
microenvironment as either deep seepage or dissolved in runoff water.
Nitrogen usually enters the soil system either in the organic form as
crop residue (and manure), as mineral fertilizers, in rainfall, and
from the atmosphere via nitrogen fixing bacteria on the roots of
legumes.  It is removed from the soil system primarily either as
ammonium or nitrate ions in runoff, deep seepage or drain effluent, as
N2 gas after denitrification by specialized soil bacteria, or as
organic nitrogen contained in crops which are harvested (Russell 1961).
The loss to runoff, drain flow and deep seepage can be considered as a
loss to the environment external to the plot.  It can be considered as
undesirable from the standpoint of aquatic nutrient control and ground-
water quality.   This study has made it possible to determine the
relative magnitudes of these losses, and the effect of various
parameters on them.
                               50

-------
    0)
    M
    O
    o
    O
    CO
    O
    01
    60
    cfl
    a
    a)
    01
    en

    (X
    
-------
Ul
KJ
                       cd

                       C


                      2
                          25%
                          20%
                          15%
c
QJ
O


«  10%
                      M-l
                           5%
                           0%
                                       poor management



                                       good management
                                         M
A     S     0

      Months

                                                       N
M
      Figure 28.  Monthly  Averages  of the Percentage of Rainfall that Occurred as Surface Runoff Under Two


                                             Management Systems, 4/70 - 3/71

-------
          •H
          r£>
          Cd
          oo
          Q)
          n
          so
          60
            C
            cu
            o
            vi
            0)
            ex
               60
               40
              20
                                         good
                                       management
                                                           poor
                                                         management
                      Good  Poor

                      Management
                                       Low   High        Low

                                               Fertility
High
Figure 29.
             Percent Aggregate Stability of Soil on Plots Under Different
                    Residue Management and Fertility Levels - 4/70-3/71
               Deep  Seepage and  Surface Runoff Losses

Knowledge of subsurface water nitrate and ammonium concentration has
enabled calculation  of the magnitude of this deep seepage loss of
nitrogen to the environment.  It has been found  that ammoniacal loss in
deep seepage is negligible, but  that nitrate loss may reach as high as
200 Ibs./acre/year  (see Appendix B, Table 5).  By comparison, surface
runoff loss of nitrogen, both dissolved and contained in suspended
sediments, was usually small except where heavy  rainfall followed recent
fertilizer application (see Figure 30) and where management conditions
were least effective in controlling runoff, i.e., no plant cover, ard
low soil aggregate stability.  During the period of mid-October to n.id-
November, the total  rainfall was 6.37 inches.  The relatively large loss
of nitrate-nitrogen  under high fertility beans and poor management can
be attributed to this one month  period.
                                53

-------
Figure 31 compares surface runoff for each treatment.  Differences
between crops were slight and reflected mainly by the amount of ground
cover provided by the crop.  The slight difference between fertility
levels apparently reflects the effect of fertilizer on the soil physical
conditions as previously discussed.  A highly significant difference in
runoff can be attributed to management.  Good management practices
(return of crop residue and/or cover crop) had the effect of reducing
runoff by 50 percent.

The nitrogen and phosphorus losses in runoff (Figure 32) are correlated
with the quantity of surface runoff and fertilizer application rate.
This implies that time of fertilizer application is important.  Indeed,
it is critical, for as Figures 33, 34, and 35 show, the deep seepage
loss is also affected by time of application.  Without exception, the
greatest loss to deep seepage was found to occur in November.  When no
fertilizer was applied this loss was not of much consequence.  However,
where heavy fertilization occurred in October (as for example the bean
plots in preparation for wheat, see Table 1, Appendix B) then the nitrate
loss in the deep seepage was clearly excessive.

In general, the deep seepage loss of nitrogen is the major component of
losses of nitrogen to the environment.  They may be of considerable
concern in regions where subsurface drainage is widely used to remove
excess water from the soil profile.  It can be seen from this study
that the type of crop and the type of management do not have much effect
on subsurface loss of nitrogen, but that rate of fertilizer applied has
a very marked effect.  This effect is non-linear in nature, and is
undoubtedly greatly increased as fertilizer application rate is in-
creased (see Figure 36).  Thus, if less than 100 Ib. of nitrogen
fertilizer is added in the spring, after deep seepage stops in May,
and if runoff is kept to a minimum by good cover and conservation
practices, then the loss of nitrogen does not appear to be any greater
than that from plots on which no fertilizer was added (see Appendix B,
Table 5).

Figure 37 shows the subsurface concentrations of nitrate nitrogen
averaged for the hydrologic year.  If 10 P.P.M. is the limit of nitrate
plus nitrite nitrogen permissible for domestic water supply (FWPCA,
1968), then it can be seen that if available nitrogen (total inputs)
exceeds 175 Ibs./acre, there is a high probability that subsurface water
flow would be unacceptable for domestic supply on the criterion of
nitrate alone.  Thus, fertilizer application would need to be restricted
to approximately 125 Ibs./acre where no crop residue is returned, or
100 Ibs./acre where residues are returned (depending on type of residue),
to meet this water quality standard (Appendix B, Table 5).  By contrast,
surface water only exceeded permissible nitrate levels for domestic
supply on one treatment - high fertilizer application'(490 Ibs./acre)
in the fall with poor soil cover and residue management.

Nitrogen from other sources has been calculated, namely, the inorganic


                               54

-------
Ui
Ul
               M-4
               4-1
               O
               a
               V*4
4J
CO  CO
O  ,O
                n
               O
12


10


 8


 6

 4
                                                                    Beans/wheat, high fertility,
                                                                          poor management
                             M
                                 A     S     0      N

                                    Months (1970-1971)
                                                    D
M
 1 other
treatments
less than
1 Ib./A
       Figure 30.  Nitrate-Nitrogen Lost in Runoff Water by Months

-------
                  140
              o
              o
              o
X
cu
    130
                  120
              to
              c
              o
              cd
              O
              4-1
              4-1
              O
              c
              Pi
              0)
              a
              cd
              4-t
              M
              3
    110
                  100
     90
     80
                   70
                   60
                           Corn 'Beans Wheat'  Normal  High    Good  Poor
                                Wheat
                               CROP
                                FERTILITY
MANAGEMENT
Figure 31.  Total Accumulative Annual Surface Runoff Volumes - 4/1/70 -
            3/31/71.

nitrogen released by the soil organic matter (mineralization) and that
contained in organic matter returned to the soil as crop residues.  It
does not appear that the source of the nitrogen has any effect on loss,
except in that management (amount of crop residue returned and crop
cover) affec.ts runoff and soil aggregate stability significantly  (see
Figure 29).

Application times for non fertilizer nitrogen inputs are more difficult
to predict.  They are influenced by temperature, moisture, carbon/
nitrogen ratio and all other factors which customarily affect the rate
of bacterial growth and decay functions.  It can be seen that nitrogen
applied as mineral fertilizer is readily soluble and thus easily  lost
                               56

-------
            7
/
                               2
Legend

N03-N
PO.-P
  4
                                               /
                       X
                       X

Corn    Beans  Wheat
        Wheat

       CROP
                                                                       C/5
                                                                       O
                                                                       h-1
                                                                       c
                                                                       a4
                                                                       O
                                                                      2*-,
                                                                       a4
                                                                       CO
                                    High
         Normal
                                           Good
                            Poor
                                      FERTILITY        MANAGEMENT
Figure 32.  Total Accumulative Annual Nitrate-Nitrogen Phosphorus  (Solu.
            P04-P) Losses  from Surface Runoff  - 4/1/70 - 3/31/71.
in seepage or surface runoff where  these occur immediately after appli-
cation.  However, nitrogen in crop  residues and other organic forms
(e.g., manure) may not be released  so rapidly.  Rapid release could take
place only if conditions for bacterial growth are optimum.  Thus, high
carbon residues and manure applied  with nitrogen fertilizer may tie up
some nitrogen in cell material and  act as a "buffer" to prevent
excessive losses during adverse weather conditions.  Further studies on
this same area are proposed to help determine to what extent manure
nitrogen is retained during high deep seepage periods in comparison to
fertilizer nitrogen.
                                57

-------
                               (U
                               60
                               cd
                               p-
                               cu
                               OJ
                               CO
                               0)
                               a)
                               Q
00
                               CO — .
                               CO •
                               o co
                               a)
                               W>
                               O
                               ro
60
                                   50
                                 2 40
                                   30
                                   20
                                   10
A - High fertility with  residues

B - High fertility without  residues

C - Normal fertility with residues

D - Normal fertility without  residuesy
               Corn
                                    Figure 33.  Monthly Nitrate-Nitrogen Loss to Deep  Seepage - Corn -

                                                             4/1/70 - 3/31/71

-------
               90
               80   -
               70
               60
               50
               40
               30   "
               20   -
               10   -
                °AMJJASON

                                                  Months


Figure 34.   Monthly Nitrate-Nitrogen Loss to Deep Seepage - Beans/Wheat  _


                               4/1/70 - 3/31/71
M
cfl
&

-------
  30
0)
60
cd
P» on
a; zu
(D
         A - High fertility with residues
         B - High fertility without residues
         C - Normal fertility with residues
         D - Normal fertility without residues
                  Wheat
          M     J     J     A     S     0  •'   N     D
                                  Months

Figure 35.   Monthly Nitrate-Nitrogen Loss to deep seepage - Wheat  -
                        4/1/70 - 3/31/71
The actual concentration of nitrates in ground water will depend on the
dilution factor of low nitrate water from other parts of the watershed,
or ground water contributing area.  In an area such as central New York,
the proportion of land which is under high fertilizer cash crops is
small, and ground water is liable to be low in nitrates because of the
dilution effect of low nitrate water.  However, in the mid-West, where
large areas are almost entirely cultivated to cash crops, and where
ground water is shallow, nitrate concentration in ground water may
easily exceed permissible levels.  Schmidt (1956) found that well water
in Minnesota had as high as 120 ppm nitrate-nitrogen on farms where the
well was located close to barnyards or cultivated soils.  The lowest
concentrations (less than 1 ppm) occurred in wells located such that
the surrounding soils were in sod.  Tile drain effluent tended to be
                               60

-------
                 Corn

             O  Beans/Wheat

                 Wheat
                                                     Beans/
                                                     Wheat
                  100      200      300      400      500

                      Total Nitrogen Inputs Lbs. N/Acre
                                                   600
Figure 36.
Apparent Relationship Between Annual Nitrogen Inputs and Losses  to
the Environment (Surface Runoff and Deep Seepage) -
              4/1/70 - 3/31/71
                                61

-------
    f ^
    t
       70
       60
       50
       40
     fO
    i
       30
       20
       10
  9 Deep Seepage
  O Surface Runoff
                          'l  i
                  100
               200
                                Beans/wheat,   ^ Q
                             high fertility,  poor
                                management
300
400
500
600
                    Total Nitrogen Inputs (Lbs. N/Acre)
                  (Fertilizer + Mineralization + Residue)
Figure 37.
Annual Relationship Between Total Nitrogen Inputs and
Nitrate N Concentrations in Subsurface Flow (Deep Seepage)
and Surface Runoff - 4/1/70 - 3/31/71
                               62

-------
intermediate with average concentrations of 15-20 ppm.  Samples were
collected in the summer.  These high  (FWPCA, 1968) levels of nitrate
nitrogen are most likely to be a result of high rates of nitrogen ferti-
lizer applications in these areas.

The desirability of high nitrogen applications is also questionable
from an economic standpoint.  Crop yield and uptake of nitrogen was not
higher under high fertility conditions except under poor residue manage-
ment practices.  Thus the suggested overall available nitrogen limits,
and fertilizer limits, would appear also to be justified on the basis
of yield response and economic return.  Unfortunately it is not possible
at this time for a dollar value to be placed on a pollution hazard.
Thus economists tend to ignore the environmental degradation which
should be considered as resulting from high fertilizer applications.
At this time, recommendations for high nitrogen applications on the
basis of projected yield returns should be carefully reconsidered.

                           Denitrification

Nitrogen is also lost from the soil as N2 gas released by certain
specialized soil bacteria.  This capacity is indicated to be dependent
on the amount of nitrogen available to these bacteria (see Appendix B, -
Table 5).  Under conditions of limited nitrogen, this loss is un-
desirable, but where the soil is being used as a disposal medium then
maximum denitrification is desirable.  Further studies are proposed to
evaluate the degree to which the soil denitrifying ability can be
utilized to remove nitrogen from organic wastes.

                             Phosphorus

Phosphorus, unlike nitrogen, is not mobile within the soil and phosphate
ions do not leach readily.  Phosphorus is held tightly as a complex
anion by clays and the amount of phosphate in solution in the soil
water at any one time is extremely small.  In most cases, this will be
less than a pound per acre.

The relation of soluble phosphate to organic phosphorus and fixed
phosphate is shown below:

                                     Mineralization
                                  z_
   Inorganic ^	  Soluble     	     Organic
   Phosphorus"1      7 Phosphorus      Immobilization         Phosphorus

Soluble phosphate is in chemical equilibrium with fixed phosphorus.
When additional phosphorus is added to the soil as in the form of
chemical fertilizers, the equilibrium tendency is towards the left.
When phosphorus is removed from the soil by plant absorption, the rate
of replacement to the soluble or available phosphorus reserve is faster
with the addition of phosphorus fertilizers.  As organic phosphorus
decomposes, the soluble phosphorus reserve is temporarily restored, but


                               63

-------
the equilibrium tendency is to drive the reaction to the left.

The soil then, has the potential to adsorb and maintain large amounts
of phosphorus.  The rate of release of phosphorus is not well corre-
lated with fertilizer additions and the soil can be thought of as a
large storage area for incoming phosphorus.

Data present in Appendix B, Table 5 shows that the greater the ferti-
lizer addition, the greater will be the amount of phosphorus that will
be fixed.  Values ranged from an addition of 17 pounds of phosphorus
added and 3 pounds fixed to 139 pounds added and 126 pounds fixed.

Figure 38 (% of fixation) illustrates that the percentage of phosphorus
fixed increases as phosphorus inputs increase.  This is true because a
nearly constant amount of phosphorus is taken up by a crop of equal
yield regardless of phosphorus ..additions ,as long as the minimum
addition satisfies plant requirements.

Most of the phosphorus that is lost from the soil system to the environ-
ment is attached to the sediment as a fixed complex (Appendix B, Table
5).  Sediment influxes to watercourses is considered by many to be
instrumental in inducing algal growth.  There is much controversy as
to just how much of this phosphorus is made available to aquatic plant
growth.  Proposed research at Cornell can possibly answer this most
important question.

Any means of controlling soil losses is effectively controlling phos-
phorus losses.  Good management systems in which the plant residue is
returned to the soil promotes a better soil structure and subsequently
a decrease in soil erosion.  In addition, a cover crop grown in con-
junction with a cash crop is influential in reducing soil losses by
increasing infiltration and physically holding the soil intact.  In all
but one instance (an unexplainably large soil loss from snow melt in
March under a system of good managed corn) well managed systems had a
lower discharge of sediment in contrast with poorly managed ones.
Largest losses of sediment phosphorus (with one exception) occurred
under a system of poorly managed wheat plots.  These plots were
harvested in August and the poorly managed ones were exposed as bare
ground.  This subjected these soils to a greater amount of soil erosion.

In conclusion, the soil is a natural sink and can fix large quantities
of phosphorus.  For this reason, very small amounts of phosphorus are
present in water discharges from the land surface.  Essentially, all of
the phosphorus that is removed from a soil system is either by crop
uptake or soil erosion.  Soil erosion can be diminished by practicing
sound conservation practices such as increasing soil aggregation by the
addition of organic residues and providing surface covers in the form
of a cover crop to reduce the impact of rainfall and the dislodgement
of soil.
                              64

-------
     M
     O
    J3
     CU
     to
     O
    JS
    FM
     o
    •H
90
       75
60
     I  45
     O
     (U
     o
     M
     (U
    P-f
       30
       15
          0              50            100            150

                           Phosphorus Inputs (Lbs. P/Acre)


Figure 38.  Percentage of Phosphorus Immobilized by the Soil and

               Phosphorus Losses - 4/1/70 - 3/31/71
                                65

-------
                            SECTION VI

                        ALGAL NUTRIENT STUDY

The awareness that increased phosphorus  input  is a major factor in the
accelerated eutrophication of lakes has  resulted in the need for means
to assess the phosphorus status  of natural waters particularly as it
relates to algal growth.  The concept of a critical concentration of
phosphorus in the surrounding medium as  being  a reliable measure of
potential algal growth has received much attention.  This critical
concentration is the  concentration of phosphorus in the growth medium
above which there is  no increase in growth as  a result of further
additions of phosphorus.  Below  the critical concentration, the algae
are in a state of phosphorus limitation  and growth is strictly a
function of phosphorus concentration.  The experimental value obtained
for the critical concentration of phosphorus is highly dependent on the
type of culturing system employed and the assumptions and interpre-
tations that are made.  Azad and Borchardt (1970) using turbidostatic
techniques found this phosphorus concentration to be about 16 yM for
Scenedesmus and Chlorella.  Thomas and Dodson  (1968) using batch
cultures of Chaetocerous gracilis obtained a value of 0.22 pM.  Rhee
(1971) also using batch cultures, obtained a value of about 0.5 ]M
phosphorus for Scenedesmus.  A major disadvantage of the use of
turbidostats and chemostats to determine the critical phosphorus concen-
tration is the great  difference  which often exists between the phospho-
rus concentration in  the input and that  which  is present in the growth
vessel.  This disparity is particularly  crucial at low concentrations
of phosphorus.  Growth parameters in these systems are often based on
input concentrations. Relating  growth characteristics to phosphorus
concentrations in batch cultures is also open  to question, since the
external concentration continually decreases as growth progresses.

In an attempt to circumvent some of these problems, a culturing system
was devised which permitted the  external phosphorus concentration in
the growth vessel to  be maintained at any desired level.  Using this
type of system, the growth of Chlorella  pyrenoidosa was measured at
different maintained  concentrations of phosphorus and the results
compared with those obtained by  other workers.
                               67

-------
                       Materials and Methods

The organism used for these studies was the green alga, Chlofella
pyrenoidosa, strained 395 (Culture Collection of Algae, University of
Indiana, Bloomington, Indiana).  Stock cultures were maintained on agar
slants of the basic culture medium in screw cap culture tubes.

The basic culture solution had the following composition:  KNOs, l.SroM;
Ca(N03)2, O.SmM; MgSOi^, 0.2mM; KHC03, 0.6mM; Fe-EDDHA (ferric chelate of
ethylenediamine dihydroxyphenylacetic acid), 10 yM; HaBOs, 50 yM; MnSOt^,
1 yM; ZnSOi*, 0.7 yM; CuSOi*, 0.1 yM; Na^oO;*, 0.1 v>,; Co(N03)2, 0.1 yM.
Phosphate variables were added as aqueous solutions of KI^PO^.  The
solution was sterilized by passage through a membrane filter (0.22 y
pore size, Millipore Crop. Bedford, Massachusetts) into autoclaved
culture vessels.  The cultures were stirred by bubbling with an air-C02
mixture which had been filtered through Dacron wool and a Millipore
filter (0.22 y).  The pH of the culture solution was maintained generally
between 7.2 and 7.5 with a C02  concentration in the mixture of 1% (v/v).

The culture vessels consisted of 9-liter "Pyrex" bottles filled with 7.5
liters of nutrient solution.  These were positioned between two fluo-
rescent light banks and received continuous illumination at an intensity
at the outer surface of the bottles of 1100 ft-c.  The temperature of
the cultures was maintained at 26 ± 1 C.  Daily aliquots of 100 ml were
aseptically withdrawn from each culture.  This aliquot was divided into
two portions, one of which was used to determine algal growth, and the
other was filtered and used to  determine phosphate.  Checks were made
for bacterial contamination by streaking a portion of the culture on a
medium consisting of nutrient solution supplemented with glucose and
yeast extract and solidified with agar.

Based on the growth rate and phosphate-usage rate, the calculated amount
of phosphate to be used in the next 24 hours was added to the cultures
daily.  This was accomplished by autoclaving the necessary amount of
phosphate in a volume of 100 ml and then slowly adding this solution to
the culture by pumping with a multi-veined peristaltic pump at a constant
rate of 100 ml/24 hours.  This procedure maintained both the total
volume of the culture and the phosphate concentration in the aqueous
phase.

Growth was measured by diluting a suitable aliquot of the culture with
KCl to give a final concentration of 0.1M KC1 and counting with an
electronic particle counter (Coulter, Model B, Coulter Electronics, Inc.,
Hialeah, Florida).  Counts were taken over the particle diameter range
of 2 to 8 y  in 1-y intervals.  This range was found to include all of
the observed cell sizes of this organism.  The interval counts were
converted to cell volumes and these volumes summed for all six intervals
to obtain the total volume of cells.  Growth rates were calculated for
the exponential growth phases using the expression [Iog2 (Vol2 / Voli)]
x (t2 - ti)  1.   The reciprocal of the growth rate thus defined is the
doubling time.
                                68

-------
Phosphate was determined on the filtered  (0.22  u ) solutions by the
method described by Murphy and Riley  (1962).  Discarding the first
20 ml of solution to pass through the filter was necessary to avoid
phosphate contamination from  the filter.  Cellular phosphorus was
determined by dry-ashing the  cells and determining total phosphorus
(Greweling, 1966).

Low phosphorus inoculum was used for  all  experiments.  Cells were
allowed to grow for several days in low-phosphate liquid media to
insure a low-phosphorus status.

All treatments were replicated  three  times.  Occasionally a culture
would develop bacterial contamination or  other  problems would arise and
this culture would be  discarded.  Accordingly,  the values presented are
the means  of  three observations  generally,  but  only  two observations
occasionally.
                                 69

-------
                      Results and Discussion

The growth of Chlorella pyrendiddsa at maintained concentrations of
phosphate can be separated into three phases.  These phases can be
easily recognized in Figure 39.  The first is characterized by a rapid
growth rate which lasts for about one day after inoculation and most
likely represents an adjustment to the new physical environment.  This
is followed by an initial exponential phase of growth which is slower
than the final or adapted exponential phase.  The duration of the initial
exponential phase is dependent on the phosphate concentration of the
nutrient medium, being 0 and 6 days at 10 and 0.1 pM phosphate,
respectively.

Table 5 shows the growth parameters of the initial and final exponential
phases at various phosphate concentrations.
Table 5.  The Effect of Various Maintained Phosphate Concentrations on
          the Initial and Final Exponential Growth Rates of Chlorella
          pyrenoidosa.
Phosphate
Concentration
yM
0.1
1
10
Initial Exponential
Growth Phase
Doubling time
days
1.96
1.45
—
Growth rate
day'1
0.51
0.69
—
Final Exponential
Growth Phase
Doubling time
days
1.03
0-85
0.86
Growth rate
day'1
0.967
1.17
1.16
At 10  yM phosphate, no initial exponential growth phase is evident and
the culture passes directly into the final exponential growth phase with
the volume of cells doubling every 0.86 day.  When the maintained
concentration of phosphate is decreased to 1 yM, an initial exponential
growth phase lasting about three days is observed but the final
exponential growth rate is the same as that observed with 10 yM
phosphate.  A further decrease in the phosphate concentration to 0.1 yM
results in a longer initial exponential growth phase at an even slower
growth rate.  However, the final growth rate at 0.1 yM phosphate is
0.967 day 1 compared to 1.17 and 1.16 day 1 and 1 and 10 yM phosphate,
respectively.
                               70

-------
 &

1
 p.
 o
 I-J
 w
 u
                                                    TIME  (Days)


   Figure 39.     Effect of Various Maintained Phosphate  Concentrations  on  the  Growth of Chlorella

                  pyrenoidosa.  Curves are Displaced 3 Days for  Ease of  Interpretation.

-------
Several observations may be made from this data.  It can be noted that
at concentrations below 10 yM phosphate, an initial exponential growth
phase is observed which has a growth rate of approximately one-half
that of the final growth .rate.  This phenomenon undoubtedly represents
a cellular adaptation to low external concentrations of phosphate.
Another important observation is that a maintained phosphate concen-
tration of 1 yM will support maximum growth of this alga.  The actual
critical concentration of phosphate is between 0.1 and 1 yM.  This value
is in general agreement with the values obtained for other algal species
[Rhee (1971); Thomas and Dodson (1968)] using batch cultures.  Experi-
ments currently in progress will measure the growth rate of this alga in
batch culture, but the observed agreement of the results of the
maintained phosphate experiments with the batch culture experiments of
others suggests that batch culture data may give reliable results in
determining critical phosphorus concentrations.  The final exponential
growth rate at the very low phosphate concentration of 0.1 yM is still
over 80 percent of the maximum growth rate observed in these experi-
ments.  This indicates that the external concentration of phosphate must
be extremely low (well below 0.1 yM) before growth of this organism is
severely curtailed.

From the discussion thus far, one might conclude that due to the low
critical phosphate concentration for this alga, control of such an
organism in a natural system would be nearly impossible.  There are
several important factors which must be considered, however.  First of
all, it is important to understand that the growth rates obtained in
these experiments were with maintained concentrations of phosphate.  In
order to support a sustained growth rate at very low concentrations of
phosphate, the total amount of phosphate that must be added is very
great.  It thus becomes necessary to consider the total amount of
phosphate that must be supplied to support a given standing crop of
algae.

A preliminary batch culture experiment has shown that the maximum cell
volume attainable with this organism is 6.7 x 106 and 6.2 x 107  y?ml
from initial phosphate concentrations of 1 and 10 yM, respectively.  In
this experiment a two-liter bottle containing nutrient solution with
1 yM or 10 yM phosphate was inoculated with phosphorus-deficient
inoculum.  Growth was followed until the culture reached a maximum cell
density and growth ceased.  In both cases, analysis of the cell-free
solution showed that there was no phosphate remaining in solution when
growth ceased.

It was considered of interest to determine the pattern of algal growth
if the cultures were maintained at given concentrations of phosphate
and then at some point the addition of phosphate to the cultures be
discontinued.  Figures 40 and 41 show the results of two such experi-
ments.  Following the cessation of phosphate addition, the cell volume
doubled at both the 0.1 and 1.5 yM levels.  A much more rapid and
complete removal of soluble phosphate occurred at the 1.5 yM level than
                               72

-------
.£!


i

•3-
.J
o
d
w
o
       5x10
       10
        5x10
       10
                                                                                          .  0.12
                                                                                          -  0.08
                                                                                                   is

                                                                                                   O
                                                                                                   o
                                                                                                   CO
                                                                                          -  0.04
               -2
            Figure 40.
                            -1
+4
                                      TIME (Days Relative to P Cessation)


                          The Effect of Eliminating Phosphorus Supply on the Solution Phosphate

                          Concentration and Growth of Cultures of Chlorella pyrenoidosa previously

                          Maintained at Approximately 0.1 yM Phosphate.

-------
 a


T

 3-
w
o
      5x10
          8
      10
      5x10
      10
             -2
                         -1
    0           +1          +2


TIME (Days Relative to P Cessation)
+3
+4
                                                                                         1.5
                                                             s

                                                             ^/




                                                        i.o  S
                                                             H



                                                             to
                                                                                      .  0.5
                                                                                         0
           Figure 41. The Effect of Eliminating Phosphorus Supply on the Solution Phosphate

                      Concentration and Growth of Cultures of Chlorella pyrenoidosa Previously

                      Maintained at Approximately 1.5 yM Phosphate.

-------
at the 0.1  ]M level.  This may have been due in part to differences
between the cultures in  the density of  the algal cells in the culture
at the time phosphate additions were discontinued.  If the absolute
increase in cell volume  after  the  cessation of phosphate addition is
measured, it is observed that  the  cultures which had been maintained at
0.1 uM phosphate increased by  4 x  106 y3/ml.  The cultures which had
been maintained at  1.5 uM phosphate increased by 70 x 106 y3/ral.  The
data from the batch culture experiment  would have predicted increases
in cell volume about one-sixth of  that  observed.  The most obvious
explanation for this observed  difference is that even at very low
external phosphate  concentrations, this alga can store some phosphorus
in excess of its growth  needs  and  can utilize this stored phosphate for
growth when the external phosphate has  been exhausted.  This and other
possible explanations for the  disparity between batch culture and
maintained culture  experimental results are currently being investigated.

Another approach to predicting maximum  algal growth is the use of
minimum phosphorus  concentrations.  For an algal cell to continue to
grow and carry out  is essential metabolic functions, it requires a
certain minimum internal concentration  of phosphorus.  This minimum
internal phosphorus content should provide the basis for calculation
of the theoretical  maximum algal volume that could be obtained from a
given initial concentration of phosphate in the surrounding medium.
The calculated value for the batch cultures is 1.6 x 10 10 ymole/y3 of
cell material.  Rhee (1971) found  a nearly identical value of 1.56 x
10 10 ymole/y3 for  batch cultures  of the green alga, Scenedesmus.  An
attempt was made to measure this value  directly using cells that had
been grown previously at maintained levels of phosphate.  One liter of
the cultures depicted in Figure 39 was  centrifuged when growth had
ceased and the cells analyzed  for  total phosphorus.  The cells from
this liter contained an  average of 217  yg of phosphorus.  This converts
to a minimum cellular phosphorus concentration of 0.5 x 10 10 ymole/y3.
This value is about one-third  that obtained from calculations using the
batch culture data. This procedure is  subject to errors both from cell
loss during centrifugation and analytical problems due to the small
sample size.  It is doubtful,  however,  that these errors could totally
account for the observed discrepancy.   An explanation for the difference
between batch and maintained cultures in this respect is also currently
being sought.

Using the value of  1.6 x 10~10 ymole/y3 as a minimum phosphorus concen-
tration, one can calculate the maximum  algal volume which could be
produced by the addition of various amounts of phosphate to a given
volume of a phosphorus-deficient algal  culture.  Figure 42 relates this
expected algal volume to the amount of  phosphate added per liter of
culture.  It should be emphasized  that  these are maximum values since
all of the phosphorus is incorporated into cellular material and none
of the cellular material is lost due to sedimentation.  Considering that
it requires an algal cell volume of about 107 y3/ml to produce a
moderately severe water  quality problem, it can be seen from Figure 42
                               75

-------
      10° .,
      10  .
w
PH
i
      10
      10
          0.1


       Fieure 42.
                     1
       PHOSPHATE ADDED (ymole/liter)

Relation Between Phosphate Addition and Maximum
Expected Algal Cell Volume.
10
                                76

-------
that to produce an algal population of  this magnitude approximately
1.5 ymole of phosphate per liter  is required.  Although cell volumes
of 2 x 10  y /ml can be produced  with a phosphate concentration
maintained at 0.1 yM, reference .to Figure 40 shows that an unreplenished
phosphate concentration of 0.1 yM will  produce a maximum growth of
only 6.5 x 105 y3/ml.  This means that  to produce a cell volume of
10  y /ml with an initial phosphate concentration of 0.1 yM, a replenish-
ment factor of at least 15 is required.  It thus becomes evident that to
limit algal growth in a lake or pond situation, it is necessary not only
to maintain low concentrations of phosphate in the aqueous phase, but
also to minimize the total amount of phosphate x^hich enters the body of
water.

Another factor which must be considered in a natural situation is the
limitation of algal growth due to such  processes as competition for
nutrients by other organisms and  grazing by organisms higher in the
food chain.  Khee  (1971) has shown that competition by bacteria for
phosphate can be significant in influencing the phosphate-dependent
growth of Scenedesmus.

Some final factors which also must be  considered are:  loss of phosphate
to  and release from bottom sediment during the growing season, and
supply of phosphorus from sources other than phosphate such as soluble
organic phosphorus compounds and  particulate forms of phosphorus.  In
lakes which develop a  thermocline during  the summer months, transport
of  any phosphates  released from the sediments may be too slow to be of
much significance  and  indeed Fitzgerald (1970) has found aerobic lake
muds to be particularly poor sources of phosphorus for Selenastrum.

Probably one  of  the biggest uncertainties at present in predicting the
crop of algae which a  given lake  will  produce centers around the
extent of which  incoming soluble  organic  and particulate forms of
phosphorus become  available for algal  growth during the growing season.
                               77

-------
                            SECTION VII
                      ECONOMIC CONSIDERATIONS

The soil and water conservation movement started in  the United States
in the early 1930's.  As a  result  of  research and practical observations,
many findings are available on economic benefits of  soil and water
conservation to individual  farmers (Anonymous,  1956; Free, 1956; Free,
1970; Sauer, McGurk  and Norton, 1950).  These benefits have also been
passed on to groups  of people.  This  has happened through soil and water
conservation districts, watershed  groups,  and larger valley compacts.

The newer environmental concepts of water  quality in relation to animal
waste, fertilizer, and land management have not been fully appreciated.
Research findings may still be too few to  fully apply these concepts to
the present land  and water  situation.  However, it would seem in order
to point out existing benefits to  farms from improved management of
manure, fertilizer,  and the land resource.  Emphasis here is placed upon
the findings of this research.  The discussion will  be concerned with
the individual farm.

In a previous study  (McEachron, et al., 1969) it was suggested that on
farm studies of costs of hauling and  spreading manure for dairy were
$2.19 per ton.  This same study also  showed that under field conditions
of good management,  dairy manure when applied to corn would yield a
return of $1.42 per  ton.  This income return would be obtained from the
production increase  of continuous  corn silage.

Using data presented in this report,  Coote (Coote, 1971) calculated a
return of 75C per acre for  reduced erosion losses due to manure appli-
cation on continuous corn.   Manure only and fertilizer plus manure
increased nitrate losses at 20 year storm  intensities only.  Up until 10
year storm frequencies values of nitrates  lost  from  manured and non-
manured treatments were the same.   Orthophosphate values in runoff were
reduced by manure.   Manure  on continuous corn,  either alone or with
fertilizer, reduced  losses  of organic matter, nitrogen and phosphorus by
50%.  Nitrogen and phosphorus are  priced at approximately IOC per pound
in fertilizers.   If  the assumption is made that the  organic nitrogen and
phosphorus retained  on the  land is worth the fertilizer price, then it
is possible to add IOC more to the benefits of manure.

Table 6.  Cost and Return per Ton  of  Manure Spread on Continuous Corn
          Silage  Under Good Management
Labor
Power
Equipment
Total
Net
Costs
.91
.61
.67
$ 2.19
Returns
Silage yield increase
Conservation benefits
Fertility saved
1.42
.75
.10
$ 2.27
.08
                                79

-------
It should be pointed out that the benefits calculated for continuous
corn culture in terms of erosion control and fertility kept on the
land could not be shown in terms of mixed cropping and continuous hay
studies.  This is true because the influence of the close growing crop
in the rotation (alfalfa) tended to improve the physical condition of
the soil to the extent that the beneficial effects of manure were no
longer evident.  However, (McEachron, et al. , 1969) yield increases on
alfalfa were 5% for each ton of manure applied.  Calculated as haylage,
this alfalfa yield increase is valued at $2.88.  This is well in excess
of the cost of hauling and spreading one ton of manure.  -'•>-

The field scale water study suggests that in exceeding normal rates of
fertilization, farmers may not only pollute ground and surface waters,
but also spend unnecessarily on fertilizers.  The unnecessary expense
may range from $20 to $40 per acre per year.
                               80

-------
                          SECTION VIII


                         ACKNOWLEDGEMENTS
The support of Professor Madison J. Wright, Head of the Department  of
Agronomy is gratefully acknowledged.

The following Agronomy staff members made major contributions to writing
and preparing data for the report:  D. R. Bouldin, T.  E. Greweling,
S. D. Klausner, D. J. Lathwell, D. 0. Wilson, and P. J. Zwerman. They
were assisted by the following graduate assistants and technicians:
D. R. Coote, A. B. Drielsma, D. F. Ellis, G. D. Jones, S. Lee, J. P.
Loch, and R. Jones.

Professor R. C. Loehr and C. A. Marion contributed valuable services
and advice which was greatly appreciated.

The support of the project by the Office of Research and Monitoring,
Environmental Protection Agency, and  the help provided by Mr. Jeffery
Denit and Mr. George W. Bailey, the Grant Project Officer, is
acknowledged with sincere thanks.
                                81

-------
                               SECTION IX

                               REFERENCES


 1.    Allis, John A.  Comparison of Storm Runoff Volumes from Small,
      Single-Crop Watersheds and from a Larger, Mixed-Crop Watershed.
      Agr. Eng. 43 (4):220-223.  1962.

 2.    Amerman, C. R. and McGuinness, J. L.  Plot and Small Watershed
      Runoff.  Its Relation to Larger AReas.  Trans. ASAE. 10:464-466.
      1967.

 3.    Anonymous.  Conservation on Rented Land.  North Central Region
      Publications No. 69, Bulletin 377.  Kansas State College of
      Agriculture and Applied Science, Manhattan, Kansas.  1956.

 4.    Anonymous.  Selected Runoff Events for Small Agricultural Water-
      sheds in the United States.  U.S.D.A.  A.R.S.  Soil and Water
      Conservation Research Division, Washington, D. C.  20025 in
      cooperation with State Agr. Exp. Sta.  1960.

 5.    Azad, H. S. and J. A. Borchardt.  Variations in Phosphorus Uptake
      by Algae.  Environ. Sci. Technol. 4:737-743.  1970.

 6.    Brakensiek, D. L.  Selecting the Water Year for Small Agricultural
      Watersheds.  Transactions of ASAE, pp. 5-8.  1950.

 7.    Bryant, J. C., Bendixen, T. W., and Slater, S. C.  Measurement
      of the Water-Stability of Soils.  Soil Sci. 65:341-345.  1948.

 8.    Coote, D. R.  The Economics of Manure Use and Disposal with
      Reference to Environmental Consideration.  Term Paper. Ag. Eng.
      505 and Ag. Econ. 550 (Unpublished).  1971.

 9.    Federal Water Pollution Control Administration.  Report of the
      Committee on Water Quality Criteria,  p. 23.  April.  1968.

10.    Fitzgerald, G. P.  Aerobic Lake Muds for the Removal of Phosphorus
      from Lake Waters.  Limnol. Oceanog. 15:550-555.  1970.

11.    Free, G. R.  Investigations of Tillage for Soil and Water Con-
      servation I.  A Comparison of Crop Yields or Contour vs. Up and
      Downslope Tillage.  SSSA 20:427-429.  1956.

12.    Free, G. R.  Minimum Tillage for Corn Production.  Bulletin 1030.
      Cornell University Agricultural Exp. Sta., Ithaca, New York.
      1970.

13.    Greweling, H. T.  The Chemical Analysis of Plant Tissue.
      Agronomy Mimeo, No. 6622, Cornell University.  1966.


                                    83

-------
14.   Hall, A. R.  Early Erosion-Control Practices in Virginia.  U. S.
      Dept. Agr. Misc. Pub. 256.  1937.

15.   McEachron, L. W., Zwerman, P. J., Kearl, C. D., and Musgrave, R.
      Economic Return from Various Land Disposal Systems for Dairy
      Cattle Manure.  Proceedings of the Conference on Animal Waste
      Management, Cornell University,   pp. 393-400.  1969.

16.   Middleton, H. E., Slater, C. S.  and Byers, H. G.  Physical and
      Chemical Characteristics of the  Soils from the Erosion Experiment
      Station.  U. S. Department of Agr. Techn. Bull. 316. 51 pg.
      1932.

17.   Murphy, J. and J. P. Riley.  A Modified Single Solution Method
      for the Determination of Phosphate in Natural Waters.   Anal.
      Chem. Acta. 27:31-36.  1962.

18.   Rhee, G-Yull.  Competition Between an Alga and an Aquatic Bacte-
      rium for Phosphate.  Ph.D. Thesis, Cornell University.  1971.

19.   Russell, E. E.  Soil Conditions  and Plant Growth.  Longmans,
      Green and Co., Ltd.  Ninth. Ed.   1961.

20.   Sauer, E. L., McGurk, J. L., and Norton, L. J.  Costs  and
      Benefits from Soil Conservation  in Northeastern Illinois.
      Bulletin 540, Agr. Exp. Sta.s Urbana, Illinois.  1950.

21.   Schmidt, E. L.  Soil Nitrification and Nitrates in Water.  Public
      Health Reports.  71:491-503.  1956.

22.   Swanson, N. P.  Rotating-boom Rainfall Simulator.  Trans. ASAE 8:
      71-72.  1965.

23.   Thomas, W. H. and A. N. Dodsrr .   Effects of Phosphate  Concentra-
      tion on Cell Division Rates a .  Yields of a Tropical Oceanic
      Diatom.  Biol. Bull. 134:199--7.08.  1968.

24.   Thornthwaite, C. N. and Mather,  J. R.  The Water Balance.  Drexel
      Inst. of Techn. Climatology 8:104.  1957.

25.   Wischmeier, W. H.  Punched Cards Record Runoff and Soil-Loss
      Data. Agr. Eng. 36:664-666.  1955.

26.   Wischmeier, W. H.  A Rainfall Erosion Index for a Universal Soil
      Loss Equation.  Soil Sci. Soc. Am. Proc. 23:246-249.  1959.

27.   Wischmeier, W. H.  Cropping-Management Factor Evaluation for a
      Universal Soil Loss Equation.  Soil Sci. Soc. Am. Proc. 24:322-
      326.  1960a.
                                   84

-------
28.   Wischmeier, W. H.  Erosion from Corn after Meadow Depends on
      Quality of Sod Crop.  Crops and Soils 12(6):25-26.  1960b.

29.   Wischmeier, W. H., and Smith, D. D.  Predicting Rainfall-Erosion
      Losses from Cropland East of the Rocky Mountains.  U. S. Dept.
      of Agr. A. R. S. Handbook 282.  1965.
                                     85

-------
                               SECTION X

                         LIST OF PUBLICATIONS

 1.    Bouldin, D, R., Reid, W. S. and Lathwell, D, J.  Fertilizer
       Practices which. Minimize Nutrient Loss.  In Press.  Cornell Uni-
       versity  Agricultural Waste Management Conference (Feb., 1971).

 2.    Coote, D. R. and Zwerman, P. J.  A Conveniently Constructed Divisor
       for Splitting Low Water Flows.  Publication in Progress.

 3.    Drielsma, A. B.  A Field Study of Soil Erodibility in Relation to  •
       Agronomic Practices and Certain Soil Physical Properties.  Ph.D.
       Thesis.  Cornell University.  (Sept. 1970)

 4.    Howler, R. H. and Bouldin, D. R.  The Diffusion and Consumption of
       Oxygen in Submerged Soils.  Soil Sci. Soc. Am. Proc. 35:202-208.

 5.    Jones, G. D.  Rates and Timing of Nitrogen Fertilization in Relation
       to Nitrate Nitrogen Outputs and Concentrations in the Water from
       Interceptor Tile Drains.  M.S. Thesis,  Cornell University.
       (Sept. 1971)

 6.    Klausner, S. D., Zwerman, P. J. and Scott, T. W.  Land Disposal of
       Manure in Relation to Water Quality.  In Press.  Cornell University
       Agricultural Waste Management Conference.  (Feb., 1971).

 7.    Loch, J. P.  Soil Structural Stability of a Glossoboric Hapludalf
       Fine Loamy, Mixed, Mesic as Influenced by Crop Sequence and Soil
       Management.  M.S. Thesis.  Cornell University.  (June 1971).

 8.    McEachron, L. W., Zwerman, P. J., Kearl, C. D., and Musgrave,  R. B.
       Economic Return from Various Land Disposal Systems for Dairy Cattle
       Manure.  Cornell University Agricultural Waste Management Conference
       Proceedings.  (1969).

 9.    Swader, F. N. , Zwerman, P. J. and Klausner, S. D.  Fertilizing Crop
       Land - Water Quality.  Cropping Up  Vol. XIV, No. 8.  New York State
       College of Agriculture, Cornell University.  (1970).

10.    Wilson, D. 0.  The Growth of Chlorella pyrenoidosa at Maintained Con-
       centrations of Phosphate.  Publication in Progress.

11.    Zwerman, P. J., Drielsma, A. B., Jones, G. D., Klausner, S. D., and
       Ellis, D.   Rates of Water Infiltration Resulting from Applications
       of Dairy Manure.  Cornell University Conference on Agricultural
       Waste Management,  pp. 263-270.  (Jan., 1970).
                                 87

-------
12.   Zwerman, P.  J., Greweling,  T.,  Klausner,  S.  D.  and Lathwell, D- J-
      Nitrogen and Phosphorus Content of Drainage  Water at Two Levels
      of Fertilization.   Submitted for publication in the Soil Sci.
      Soc.  Am. Proc.

13.   Zwerman, P.  J.   Control of  Agricultural Effluents from Commercial
      Fertilizers.  Geneva Agricultural Experiment Station. Proceedings
      Agricultural Water Resources Symposium.  Geneva Agric. Exp.  Sta.
      pp 86-92.  1971.

14.   Zwerman, P.  J., Klausner,  S. D., Shell, B.,  Romanowski, R.
      Managing Fertilized Cropland for Improved Water Quality.  New
      York's Food and Life Sciences Quarterly,   pp 6-8.   Vol. 3, No.  2.
      (April-June, 1970).
                               88

-------
Chemostat


Coshocton Wheel



Cubic Micron

Deep Seepage

Divisor


Flume



Intercepter Ditch


Micron

Micromolar


Micromole

Millimolar

Millimole

Runoff Coefficient


Stage Height



Thermocline



Turbidostat
                              SECTION XI

                               GLOSSARY
A continuous-flow culture system in which
the flow rate is kept constant.

A runoff sampler that divides the flow from
an experimental area and retains a propor-
tional part of it in a storage tank.

10    cm ; abbreviation y .

Water leaving the soil downwards permanently.

Mechanical device for dividing up small
flows of water.

A device used to measure the discharge of
water.  Different types of flumes exist
which vary in design.

A ditch used to divert water towards a
given point.

10   meter; abbreviation, y.

10~  mole/liter; abbreviation, yM; ly M P =
31 yg/1 P = 31 ppb expressed as P.

10   mole/ abbreviation, ymole.
  _3
10   mole/liter; abbreviation, mM.
  _o
10   mole/ abbreviation, mmole.

The percentage of rainfall that occurred as
surface runoff.

A measure of the head in feet or inches.  It
is used to calculate the flow through a
flume or weir.

The boundary between the surface water and
deeper water of a lake due to temperature
(and hence density) differences.

A continuous-flow culture system in which
the density of cells (turbidity) in the
growth vessel is kept constant.
                               89

-------
Weir                -      Serves the same purpose as  a flume but
                           differs in design.
                              90

-------
 SECTION XII

APPENDICES
A.   Small Scale Runoff Study

     Table 1:  Description of Treatments
     Table 2:  T-Test Comparisons of Treatment Means of
               Runoff Variables per Storm Frequency -
               Continuous Corn
     Table 3:  T-Test Comparisons of Treatment Means of
               Runoff Variables per Storm Frequency -
               Eight Rotations
     Table 4:  T-Test Comparisons of Treatment Means of
               Runoff Variables per Storm Frequency -
               Hay Rotations
     Table 5:  T-Test Comparisons of Treatment Means of
               Runoff Variables for Three Equal Consecutive
               Half Hour  (1.25  in.) Storms - Continuous Corn
     Table 6:  T-Test Comparisons of Treatment Means of
               Runoff Variables for Three Equal Consecutive
               Half Hour  (1.25  in.) Storms - Eight
               Rotations
     Table 7:  T-Test Comparisons of Treatment Means of
               Runoff Variables for Three Equal Consecutive
               Half Hour  (1.25  in.) Storms - Hay Rotations
     Table 8:  T-Test Comparisons of Treatment Means for
               Variables  in Sediment from each of Three
               Equal Consecutive Half Hour  (1.25  in.)
               Storms - Continuous Corn
     Table 9:  T-Test Comparisons of Treatment Means for
               Variables  in Sediment from each of Three
               Equal Consecutive Half Hour  (1.25  in.)
               Storms - Eight Rotations
     Table 10: T-Test Comparisons of Treatment Means for
               Variables  in Sediment from each of Three
               Equal Consecutive Half Hour  (1.25  in.)
               Storms - Hay Rotations
     Table 11: Enrichment Ratios (E.R.) for Variables in
               Sediment from each of Three Equal Consecutive
               Half Hour  (1.25 in.) Storms - Continuous Corn
     Table 12: Enrichment Ratios (E.R.) for Variables in
               Sediment from each of Three Equal Consecutive
               Half Hour  (1.25 in.) Storms - Eight Rotations
     Table 13: Enrichment Ratios (E.R.) for Variables in
               Sediment from each of Three Equal Consecutive
               Half Hour  (1.25 in.) Storms - Hay Rotations
     Table 14: Analysis of Variance of Several Crop Rotations
     Table 15: Analysis of Variance of Two Continuous Corn
               Rotations
                                    96
                                   J.Q2
                                   104
                                   106
                                   108
                                   110
                                   112
                                   114
                                   116
                                   117
                                   119
                                   121
                                   123
                                   124
    91

-------
     Table 16:  Correlation and Regression Coefficients of
               Surface Runoff (Gal./A)  and Soil Loss (T/A)
               During Three Storm Frequencies
     Table 17:  Correlations of Selected Constituents in the
               Soil and Sediment Derived from the Soil and
               Enrichment Ratios (E.R.) for each of Three
               Consecutive Half Hour (1.25 in.) Storms
     Table 18:  Correlations of Percent  Solids in Runoff
               and Selected Variables for each of Three
               Consecutive Half Hour (1.25 in.) Storms
     Table 19:  Correlations Among Nitrate Nitrogen and
               Soluble Orthophosphate  (PO,)-P in the Soil
               and Their Concentration  in Runoff Water for
               each of Three Consecutive Half Hour (1.25 in.)
               Storms
     Table 20:  Content and Correlations Among Constituents
               in the Soil
     Table 21:  Content and Correlations Among Constituents
               in the Sediment
     Table 22:  Stepwise Regression of Selected Constituents
               on other Constituents in Sediment

     Analytical Determinations

B.   Field Scale Water Quality Study

     Table 1:  Fertilizer Application Rates on Corn, Beans,
               and Wheat Plots - 4/70 - 3/71
     Table 2:  Surface Flow Program
     Table 3:  Tile Flow Program
     Table 4:  Surface Flow Summary Program
     Table 5:  Nutrient Balance for Nitrogen (N)  and Phosphorus
               (P) in Pounds/Acre
     Table 6:  Total Accumulative Runoff and Nutrient Losses
               from Water Quality Research Plots for a Period
               from 4/1/70 - 3/31/71
125
126
126
127
12

128

129
134

135
140
146
149

151
                                    92

-------
                             APPENDIX A


Table 1 can be used throughout Appendix A  to distinguish between treat-
ments designated by code.  Details  of cropping, fertilizing and other
management differences  are tabulated for easy reference.

Tables 2 through 4 present data  for various parameters of runoff from
the rainfall simulation.  T-test analysis  of various comparisons are
shown simultaneously, with results  noted at 5, 10  and 20 percent (/, x
and *, respectively) probability (two tailed test) for readers' infor-
mation.  Data  is shown  for each  of  the three storm probabilities, 2,
10 and 20 years.

Tables 5 through 7 present similar  data to Tables  2 through 4, but they
are shown by half hour  storm increments, the 20 year storm being made
up of three identical consecutive 1.25 inch storms.

Tables 8 through 10 present  data for various parameters of the sediment
collected from the rainfall  simulation.  T-test comparisons are shown
for each of the three consecutive half hour storms.

Tables 11 through  13 show the sediment data, by each half hour storm
increment, comparing quantities  and percentages of each variable in the
sediment with  that in the original  undisturbed soil on the plot.  The
ratio of each  pair of data is presented as the Enrichment Ratio  (E.R.).

Tables 14 and  15 present an  analysis  of variance  for a 2 x 4 x 3
factorial and  a 2  x  2 x 3 factorial experiment.   Table 14 shows little
significance between three of the four different  crop rotations for the
various parameters.   In general, hay  rotations  (H W - HZ) tended to be
lower  in total losses of constituents  from the land surface during
three  consecutive  half  hour  storms.  Table 15  is  a comparison of two
continuous  corn experiments.  The difference  exists in the rate of
phosphorus  fertilizer.   Manure showed a definite  advantage over no
manure.

Table  16  is  a  presentation of the correlation and regression between
surface runoff and soil loss.  The strength of  the linear relationship
between  the  two is highest in the two year storm  frequency for con-
tinuous corn (R2  = 70.6%) and mixed rotations  (R2 = 82.8%).   In  these
two  cases,  soil stability (aggregate  stability)  is easily destroyed and
more  soil  is  eroded  per unit volume of water  (see regression  coeffi-
cients).   In the successive 10 and 20 year storm frequencies, much of
 the  erodable soil has already been lost,  hence the poor  correlation.
The  low  coefficient  for the two year storm for continuous hay (R  -
16.8%)  proves  that the high inherent  aggregate stability resists the
movement  of  soil by water.  The correlation coefficient  increases  as
storm duration increases because soil stability is diminishing.
                                93

-------
Table 17 shows the correlation coefficients between constituents in the
sediment and the soil from which the sediment is derived along with
the enrichment ratio (mean content of constituent in sediment/mean
content of constituent in the soil).  The correlations for organic
matter, total N, and total P are generally fairly good while for sand,
silt and clay the correlations were generally poor.  The enrichment
ratios for organic matter, nitrogen and phosphorus are generally equal
to each other for a given storm, and tend to decrease slightly from A
to C.  These calculations are calculated on the basis of three equal
half hour storms with an intensity of 2.5 inches per hour rather than
on a storm probability basis.

Table 18 is the correlation between percent solids in the runoff and
selected variables as shown in the table.  These results confirm the
considerable variation among plots and storms in percent solids.  The
amount of solids in the runoff was not well correlated with other
measured variables.  These values are for three equal half hour storms
with an intensity of 2.5 inches per hour.

Table 19 presents correlations among concentrations of NO^-N in the
soil before the artificial rain was applied and the concentration in
the runoff from three equal half hour storms (1.25 in.).  None of the
variables were well correlated with other variables.  Probably this is
a reflection of the complicated patterns of water movement within a
plot and the influence of properties of the plot on this movement.
The high concentrations of NO^-N found in the runoff from some treat-
ments are a reflection both of high NOo-N content in the soil plus
appreciable downslope movement of water within the plot.  If the depth
to firm till was shallow, if the air filled pore space above firm till
was limited, and if the surface permeability remained high, then the
downslope movement of water through the surface soil would be maximum.
This would in effect mean that maximum accumulation of NO^ in the run-
off would occur.  On the other hand, if the surface sealed or if the
depth to firm till and air filled pore space were large, then downslope
movement of water through the surface soil would be limited and hence
only limited amounts of NOv-N would accumulate in the runoff.  (Note
that there was a 12 inch baffle on the downslope side of the plots
which effectively forced the water moving downslope in the soil to
"surface" and be collected as runoff.  At first glance this would
appear to bias the experimental results; however, this is not the case
since probably downslope movement of water in the surface portion of
this profile is the rule and sooner or later this water empties into a
natural drainage ditch.  Thus the runoff collected is a sample of the
water which normally moves in this fashion).

Also shown are correlations among soluble orthophosphate in the water
extracts and in runoff.  The water extracts were not effective as a
means of predicting concentrations of soluble orthophosphate in the
runoff from the three storms.  However, the correlations among concen-
trations in the runoff from the three storms were reasonably high;
                               94

-------
that is, concentration was not greatly influenced by the duration of
the storm.  In summary, the procedure used was not particularly helpful
in predicting concentrations of nitrate and orthophosphate in the run-
off.

Tables 20, 21 and 22 a statistical study of the composition of the
sediment from the rainmaker studies in the summer of 1968 was carried
out with the objective of finding if one or more of the constituents
could be estimated reasonably well from contents of other constituents.
For example, the determinations of total N and total P are rather
expensive so if these constituents could be estimated from other
simpler determinations then a considerable saving in analytical costs
could be realized.  Shown in Tables 20 and 21 are the means and stan-
dard deviations of the means of constituents in the soils and sediments
along with the correlation coefficients among constituents.  Stepwise
regression of selected constituents on other constituents are shown in
Table 22.  The results show that nitrogen, phosphorus and organic
matter  are reasonably well correlated with each other, but contents of
these constituents are not well correlated with sand, silt, or clay.
The regression studies demonstrate that nitrogen can be reasonably well
estimated from organic matter content and that phosphorus can be
reasonably well estimated from nitrogen content.
                                95

-------
vo
        Table 1.   Description of Treatment Applications per Year.





        Continuous Corn
Treatment
No. Code
1
2
3
4
5
6
7
(Ay)
(Bw)
(Ccy)
(Aw)
(Ax)
(Az)
(Ccx)
Use Manure
(Tons/Acre)
Grain
Grain
Grain
Grain
Silage
Grain
Grain
6
0
0
6
6
6
6
Fertilizer
(Lbs./Acre)
0- 0- 0
30-30-30
30-90-60
30-30-30
30-30-30
30-30-30
30-90-60

(30)
(90)
(30)
(90)
(90)
(90)
Total Fertilizer
(Lbs./Acre)
0
120
270
120
180
180
270

-------
Table 1.  (Continued)
Mixed Rotations
Manure
Code Rotation (Tons/Acre)
D Y (2) C-C-0-A-A
Corn*
Corn
Oats
Alfalfa
Alfalfa

E5Y (1) C-0-W-A-A
Corn
Oats
Wheat
Alfalfa
Alfalfa*

F5W (1) C-0-A-W-A
Corn
Oats
Alfalfa
Wheat
Alfalfa*

F5Z (2) C-0-A-W-A
Corn
Oats
Alfalfa
Wheat
Alfalfa*

0
0
0
0
0


10
0
10
0
10


10
0
0
10
10


10
0
0
10
- 10
Fertilizer
(Lbs . /Acre)

30-30-30
30-30-30 (30)
30-30-30
0-30-30
0-30-30


30-30-30 (30)
30-30-30
15-30-30 (30)
0-30-30
0-30-30


30-30-30 (30)
30-45-45
0- 0- 0
15-45-45
0-30-30


0- 0- 0
0- 0- 0
0- 0- 0
0- 0- 0
0- 0- 0
Total Fert.
(Lbs. /Acre)

90
120
90
60
60
Mean 84

120
90
105
60
60
Mean 87

120
120
0
105
60
Mean 81

0
0
0
0
0
                                                          Mean
                               97

-------
Table 1.  (Continued)
Manure Fertilizer
Code Rotation (Tons/Acre) (Lbs./Acre)
C4W (1) C-O-A-A
Corn
Oats
Alfalfa
Alfalfa*

C4Y (2) C-O-A-A
Corn
Oats
Alfalfa
Alfalfa*

C4Z (3) C-O-A-A
Corn
Oats
Alfalfa
Alfalfa

D W (1) C-C-0-A-A
Corn*
Corn
Oats
Alfalfa
Alfalfa

12
0
0
12


0
0
0
0


0
0
0
0


10
10
0
0
10
9
30-30-30 (30)
30-30-30
0-30-30
0-30-30


30-30-30 (30)
30-30-30
0-30-30
0-30-30


30-90-60 (90)
30-30-30
0-30-30
0-30-30


30-30-30
30-30-30 (30)
30-30-30
0-30-30
0-30-30
Total Fert.
(Lbs./Acre)

120
90
60
60
Mean 83

120
90
60
60
Mean 83

270
90
60
60
Mean 120

90
120
90
60
60
                                                          Mean  84
                               98

-------
Table 1.   (Continued)

Continuous Hay
Code
Rotation
                        Manure
                      (Tons/Acre;
                           Fertilizer
                           (Lbs./Acre)
                   Total Fertilizer
                     (Lbs./Acre)
HYW
         (1) W-A-A-A-A
        Wheat
        Alfalfa
        Alfalfa
        Alfalfa*
        Alfalfa
HYZ
         (2) W-A-A-A-A
        Wheat
        Alfalfa
        Alfalfa
        Alfalfa*
        Alfalfa
                10
                 0
                10
                 0
                10
                 0
                 0
                 0
                 0
                 0
15-30-30
 0-30-30
 0-30-30
 0-30-30
 0-30-30
15-30-30
 0-30-30
 0-30-30
 0-30-30
 0-30-30
(30)
105
 60
 60
 60
 60
                                                      Mean   69
(30)
105
 60
 60
 60
 60
                                                      Mean   69
JW
        Continuous Hay
                          0-45-45
                          90
               Crop Plowed under for this experiment

               (  ) = Anhydrous Ammonia Sidedress
                               99

-------
     Table 2.  T-Test Comparisons of Treatment Means of Runoff Variables per 3 Storm Frequencies -

               Continuous  Cprn
o
o
Treatment Storm Frequency Solids
(Years) (%)
Ax vs Az


Bw vs Ay


Ccy vs Ccx


Ccy vs Ay


Ccy vs Aw


Ay vs Az


2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
.49
.47
.40
.31
.55
.40
.85
.49
.49
.85
.49
.49
.85
.49
.49
.46
.62
-.62
.35
.51
.39
.46
.62
.627
.48
.807
.48
.46
.62
.62
.23X
.43
.347
.35
.51
.39
Runoff
(Gal . /Acre)
1014
13865
38060
1769
14029
33351
3519
22177
50269
3519
22177
50269
3519
22177
50269
490
960
18423
298X
76867
29600
490
963
18423
14347
8782*
27915*
490*
963X
18423*
150*
. 7275*
24717*
298
7686
29601
Soil Loss Agg. Stability 0. M. Return
(Tons/Acre) ^ (Tons/Acre/Year)
.019
.239
.580
.032
.257
.553
.165
.441
1.023
.165
.461
1.023
.165
.441
1.023
.008
.206
.601
.009 33.3 60.1
.187
.493
.008 32.2 5.1.9*
.206
.601
.0407 30.5 38.47
.2217
.535X
.008* 30.5 51.9*
. 206X
.601X
.002* 30.5 39.57
.109*
.323*
.009 51.9 40. IX
.187
.493
2.1 4.3*


2.6 6.0*


2.2 4.0*


2.2 4.0*


2.2 4.3*


4.0 4.3X



-------
Table 2.  (Continued)
Treatment
Ax vs Az
Bw vs Ay
Ccy vs Ccx
Ccy vs Ay
Ccy vs Aw
Ay vs Az
St orm Frequency NO
(Years) (Lbs
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
.043
.: 1.054
2.852
.116
" ' '.641
1.222
.044
.494
1.364
.044
.494
1.364
.044
.494
1.364
.036
.494
1.996
3-N
./Acre)
.022
.702
3.146
.036
1.040
1.996
.054
.691
2.734
.036
1.040
1.996
.005X
.396
1.745
.022
.702
3.146
Soluble Orthophosphate (PO^)-P
(Lbs. /Acre)
.0004
.0053
.0137
.0000
.0016
.0022
.0013
.0113
.0249
.0013
.0113
.0249
.0013
.0113
.0249
.0001
.0046
.0112
.0001/
.0033
.0113
.0001/
.0046
.0112/
.0005
.0048
.0162
.0001*
.0046
.0112
.0001*
.0030
.0084
.0001
.0033
.0113
    /,X, *Significance of the 20, 10 and 5% level respectively (sign ignored).

-------
      Table  3.  T-Test  Comparisons  of  Treatment M. ans  of  Runoff  Variables  per  Storm Frequency -

                8  Rotations
o
ro
Treatment
C w vs C y
bf *T

D..W vs D1y
-L JL

C.z vs D,w
4 1

C. y vs Fcz
'i- D

E..X vs Fcw
5 5

E..X vs Fcz
5 5

F..W vs Fcz
5 5

Storm
Frequency Solids
(years) (%)
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
.66
.36
.48
.33
.32
.38
.45
.45
.42
.60
.47
.36
.46
.94
.62
.46
.94
.62
.33
.56
.47
.60
.47
.36
.33
.45
.36
.33
.32
.38
.70
.67
.49
.33
.56 /
.47
.70
.67
.49
.70 x
.67
.49
Runoff
(Gal. /Acre)
1460
9338
25556
1357
11617
35727
1061
12123
32673
130
8689
28209
1596
14369
37772
1596
14369
37772
1775
12947
33856
130 x
8689
28209
510
10196
31491
1357
11617
35727
990 /
12701
37240
1775
12947
33856
990
12701
37240
990
12701
37240
Soil Loss Agg. Stability O.M. Return
(Tons/Acre) (%) (Tons/Acre/Year;
.060
.175
.554
.041
.169
.570
.026
.158
.473
.004
.064
.351
.041
.457
.928
.041
.457
.928
.038
.262
.662
.004 / 74.5 71.0 2.3
.064
.351
.009 67.1 51.7 * 2.9
.182
.466
.041 62.9- 67.1 1.1
.169
.570
.035 71.0 78.3 1 1.1
.409 /
.795 /
.038 65.2 69.1 2.3
.262
.662
.035 65.2 78.3 x 2.3
.409
.795
.035 69.6 78.3 1 2.4
.409
.795
1.1 *


1.4 *


2.9 *


2.4 *


2.4 /


2.4 /


2.4



-------
Table 3 (Continued)
Treatment
C w vs C,y
^T ^T

D..W vs D..y


C,z vs D..W


C-y vs F z
^T -J

Eqx vs F_w
J J

E,.x vs F_z


Fcw vs F^z
j j

Storm Frequency NO^-N
(years) (Lbs . /Acre)
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
2
10
20
.031
.202
1.101
.126
1.145
4.517
.068
.513
1.685
.002
2.021
5.197
.092
.902
2.848
.092
.902
2.848
.045
.797
2.338
.002 *
2.021
5.197 /
.030
.397
1.763 /
.126
1.145
4.517
.032 /
.390
2.909
.045
.797
2.338
.032
.390
2.909
.032
.390
2.909
Soluble Orthophosphate (PO.)-P
(Lbs. /Acre)
.0004
.0022
.0060
.0008
.0075
.0212
.0001
.0011
.0032
.0000
.0002
.0010
.0003
.0031
.0079
.0003
.0031
.0079
.0005
.0042
.0084
.0000 x
.0002 x
.0010 x
.0000
.0154
.0188
.0008
.0075 x
.0212 *
.0000
.0008 /
.0054 /
.0005
.0042
.0084
.0000
.0008
.0054
.0000
.0008
.0054

-------
Table 4.  T-Test Comparisons of Treatment Means of Runoff of Variables for 3 Storm Frequencies -
          3 Hay Rotations.

Treatment
H,w vs H. z
4 4

H,z vs Jw
Storm


Frequency Solids
(years) (%)
2
10
20
2
10
20
.22
.43
.32
.37
.50
.39
.37
.50
.39
.35
.48
.39


Runoff
(Gal. /Acre)
267
4939
17973
179
4365
17585
179
4365
17585
324
8901 /
27050 /


Soil Loss Agg. Stability O.M. Return
(Tons/Acre) (%) (Tons/Acre/Yei
.003
.090
.229
.003
.095
.291
.003 62.3 68.2 2.0 0.6 *
.095
.291
.005 68.2 73.6 0.57 1.74
.165
.447

-------
     Table  4.  (Continued)
Treatment
H,w vs H.z
4 4

H.z vs Jw

Storm Frequency NOo-N
(years) (Lbs./Acre)
2
10
20
2
10
20
.004
.053
.495
.020
.074
.829
.020
.074
.829
.006
.555
1.471
Soluble Orthophosphate (PO,)-P
(Lbs . /Acre)
.0000
.0006
.0034
.0000
.0005
.0014
.0000
.0005
.0016
.0000
.0091 X
.0194 X
o
Ul

-------
Table 5.  T-Test Comparisons of Treatment Means of Runoff Variables for Three Equal Consecutive
          Half Hour (1.25 in.) Storms, Continuous Corn.
Treatment
A vs A
x z


B vs A
w y

C y vs C x
c c


C y vs A
c-7 y


C y vs A
f W
\~ w

A vs A
y z


Storm
A
B
C

A
B
C
A
B
C

A
B
C

A
B
C
A
B

C
Solids (%)
.49
.47
.37

.31
.56
.33
.85
.37
.51

.85
.37
.51

.85
.37
.51
.46
.62

.61
.35
.50
.36

.46
.62
.61X
.48
.79X
.43

.46,
.62>
.61
"5T
.23
'43/
.30'
.35
.50,
/
.36'
Runoff
(Gal. /Acre)
1014
12851
24195

1769
12260
19322
3519
18657
28092

3519
18657
28092

3519
18657
28092
490
10473

17460
298X
7388'
21914

490
10473
17460
1434/
*
7348
19133X
*
490
104 7 3X
17460
*
15°*
7125*
17441
298
7388

21914
Soil Loss Aggregate ' Annual O.M.
(Tons/Acre) Stability (%) Return (Tons/Acre
.019
.219
.341

.032
.225
.296
.165
.276
.581

.165
.276
.581

.165
.276
.581
.008
.198

.395
.009 33.3
.179
.306

.008 32.2
.198
.395
.040/ 30.5
.181
.314X
*
.008 30.5
.198
.395X
*
.002^ 30.5
.107*
.214
.009 51.9
.179

.306
40.1 2.1 4.3


* *
51.9 2.6 4.0


38. tj 2.2 4.0*


* *
51.9 2.2 4.0


/ *
39.5' 2.2 4.3


40. 1X 4.0 4.3X




-------
      Table 5  (continued)
o
-4
Treatment
A vs A
x z

BW VS Ay


C y vs C x

-------
       Table 6.   T-Test Comparisons of Treatment Means of Runoff Variables for Three Equal Consecutive

                 Half Hour (1.25 in.) Storms, 8 Rotations.
o
00
Treatment
C4W vs C4Y


D W vs D Y
-*- JL

C.Z vs D W
4 1

C.Y vs F_Z
4 5

E.X vs FSW
J -s

ECX vs F..Z
5 5

FJW vs FCZ
5 5

Storm
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
Solids (%)
.67
.30
.54
.33
.29
.39
.45
.44
.41
.60
.47
.38
.46
.98
.52
.46
.98
.52
.33
.55
.46
.60
.47
.38
33
X
.45
.32
.33
.29
139
.70
.67
.38
.33/
.55
.46
.70
.67
.38
.70
.67
.38
Runoff
(Gal. /Acre)
1460
7878
16218
1357
10260
. 24110
1061
11062
20550
130
8559
19520
1596
12773
23403
1596
12773
23403
1775
11172
20909
130X
8559
19520
510
9686
21295
1357
10260
24110
990^
11711
24539
1775
11172
20909
990
11711
24539
990
11711
24539
Soil Loss Aggregate Annual 0. M.
(Tons/Acre) Stability (7«) Return (Tons/Acre
.060
.115
.379
.041
.128
.401
.026
.132
.315
.004
.060
.287
.041
.416
.471
.041
.416
.471
.038
.224
.400
.004/ 74.5
.060
.287
.009 67.1
.173
.284
.041 62.9
.128
.401
.035, 71.0
.3747
.386
.038 65.2
.224
.400
.035 65.2
.374
.386
.035 69.6
.374
.386
71.0 2.3 1.1*


51.7* 2.9 1.4*


67.1 1.1 2.9*


78. 37 1.1 2.4*

-
69.6 2.3 2.47


x /
78.3 2.3 2.4


78.3 2.4 2.4



-------
      Table 6. (continued)
o
VD
Treatment
C.W vs C.Y
4 4
• ^T

D-W vs D.Y
JU J^

C.Z vs D-W
4 1

C.Y vs F_Z
^r J


E-X vs F..W
~J J

E-X vs FCZ
5 5

FCW vs F..Z
5 5

Storm
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
Nitrate - N
mg/1 Lbs./Acre
12.4
10.6
4.5
7.0
12.1
16.2
3.8
2.2
5.2
3.7
12.5
19.2
5.9
5.2
8.5
5.9
5.2
8.5
2.4
5.0
8.2
3.7
12.5
19.2*
8.1.
4.27
7.3'
7.0
12. 1X
16. 2X
5.5
4.7
11.4
2.4
5.0
8.2
5.5
4.7
11.4
5.5
4.7
11.4
.031
.170
.814
.126
1.019
3.371
.068
.445
1.172
.002
2.019
3.176
.092
.810
1.946
.092
.810
1.946
.045
.752
1.541
.002
2.019
3.176X
.030
.368.
1.366'
.126
1.019,
3.371'
.032/
.358
2.520
.045
.752
1.541
.032
.358
2.520
.032
.358
2.520
Soluble Orthophosphate(PO^)-P
mg/1 Lbs./Acre
.030
.021
.024
.035
.075
.072
.009
.007
.013
.003
.006
.007
.022
.023
.025
.022
.023
.025
.018
.025
.024
.003*
.006*
.007*
•°07*

.020*
.035
.075*
.072*
.014^
.015'
.025
.018
.025
.024
.014
.015
.025
.014
.015
.025
.0004
.0018
.0038
.0008
.0067
.0137
.0001
.0010
.0021
.0000
.0002
.0008
.0003
.0027
.0049
.0003
.0027
.0049
.0005
.0037
.0042
.oooox
.0002X
.0008
.0000
.0154
.0034*
.0008
.0067X
.0137*
.0000,
.0008'
.0046
.0005
.0037
.0042
.0000
.0008
.0046
.0000
.0008
.0046

-------
Table 7.  T-Test Comparisons of Treatment Means of Runoff Variables for Three Equal Consecutive
          Half Hour (1.25 inch) Storms, 3 Hay Rotations.
Treatment
H4W vs H^Z


H Z vs JW
4

Storm
A
B
C
A
B
C
Solids
\ fo)
.22
.43
.29
.37
.50
.36
.37
.50
.36
.35
.48
.36
Runoff
(Gal. /Acre)
247
4692
13034
179
4186
13220
179
4186
13220
324
8577
18149
Soil Loss Aggregate Annual 0. M.
(Tons/Acre) Stability (%) Return (Tons/Acre)
.003
.087
.139
.003
.092
.196
.003 67.3 68.2
.092
.196
.005 68.2 73.6
.160
.282
2.02 0.57*


0.57 1.74*



-------
Table 7. (continued)
Nitrate - N
Treatment
H,W vs H Z
4 4

H, Z vs JW
4

Storm
A
B
C
A
B
C
mg/1
1
1
2
5
1
5
.4
.1
.6
.4
.4
.9
5
1
5
2
4
4
.4
.4
.9
.2
.7
.4
Lbs . /Acre
.004
.049
.443
.020
.054
.755
.020
.054
.755
.006
.549
.916
Soluble Orthophosphate
mg/1
.011
.022
.028
.010
.012
.008

.010
.012
.008*
.023
.061*
.061*

.0000
.0006
.0028
.0000
.0005
.0009
(PO.)-P
Lbs . /Acre
.0000
. 0005 .
.0009'
.0000
.0091X
.0103*

-------
Table 8.  T-Test Comparisons of Treatment Means for Variables  in Sediment  from each of Three Equal
          Consecutive Half Hour (1.25 inch) Storms, Continuous Corn.
Treatment Storm
Ax vs Az A
B
C
Bw vs Ay A
B
C
C y vs C x A
B
C
Ccy vs Ay A
B
C
C y vs Aw A
B
C
Ay vs Az A
B
C
Organic Matter
% Lbs . /Acre
NS1
6.9
6.6
3.8
6.0
6.0
5.6
6.5
5.6
5.6
6.5
5.6
5.6
6.5
5.6
NS
6.8
6.5
0.0
7.6
8.3'
NS
6.8
6.5
5.2
5.7
4.2
NS
6.8/
6.5
o.o-.*
8.4'
8.2*
0.0 .
7.6/7
8.3
NS
30.3
43.3
6.2
37.6
35.7
22.7
36.6
65.7
22.7
36.6
65.7
22.7
36.6
65.7
NS
29.3
50.7
0.0
22.7
46.5
NS
29.3
50.7
7.8*
27.0,
34.0'
NS
29.3
50.7
o.ox
18.7
34. 2X
0.0
22.7
46.5
Total Phosphorus
7, Lbs
NS
.144
.125
.063
.115
.112
.109
.136
.113
.109
.136
.113
.109
.136
.113
NS
.142
.121
.000
.145,
.158'
NS
.142
.121
.104
.139
.123
NS
.142
.121
.000*
.146,
.141X
.000
.145
.158'
NS
.642
.870
.117
.714
1.305
.444
.770
1.305
.444
.770
1.305
.444
.770
1.305
NS
.552
.915
. /Acre
.000
.432
.854
NS
.552,
.915
.155
.609,
.824'
NS
.552,
.915'
.000*
.325X
.584*
.000
.432
.854

-------
Table 8. (continued)
Treatment    Storm
Total Nitrogen
           Lbs./Acre
                                                          Sand
Silt
Day
Ax vs Az


Bw vs Ay


C y vs Ccx
c

Ccy vs Ay


C y vs Aw
c

Ay vs Az


A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
NS
.368
.362
.200
.306
.310
.788
.344
.319
.788
.344
.319
.788
.344
.319
.200
.306
.310
.000
.377
.419
NS
.348
.309
.250
,360
.329
NS
.348
.329
.000
NS
.405
.000*
.377
.419X
NS
1.636 1
2.398 2
.341
1.484 1
1.855 2
1.181
1.935 1
3.677 2
1.151
1.935 I
3.677 2
1.151
1.935
3.677 2
NS
1.287 1
2.545 2
.000
.608
.365
NS
.287
.545
.208
.568,
.182'
NS
.287
.545'
.000*
NS
.ooox
.000
.608
.364
NS
6.4
8.6
3.4
5.8
1.5
0.5
3.3
6.7
0.5
3.3
6.7
0.5
3.3
6.7
NS
2.9
3.2
NS
5.1
4.2
NS
2.9
3.2
3.3
5.7
11.0
NS
2.9
3.2X
NS ,
13.6
6.8
NS
5.1
4.2
NS
35.2
31.7
23.7
38.2
36.4
29.3
31.9
35.5
29.3
31.9
35.5
29.3
31.9
35.5
NS
31.1
31.4
NS
34.8
32.2
NS,
31.1'
31.4
28.5
32.6
36.0
NS
31.1
31.4
NS
24.6
29. 8X
NS
34.8
32.2
NS
58.4
59.7
40.7
56.0
62.1
70.2
64.8
57.8
70.2
64.8
57.8
70.2
64.8
57.8
NS
65.9
65.4
NS
60.1
63.5
NS
65.9*
65.4
69.7
61.6
53.1
NS
65.9
65. 4X
NS
61.8
54.4
NS
60. 1X
63.5
       No sample

-------
Table 9.  T-Test Comparisons of Treatment Means for Variables in Sediment for each of
          Three Equal Consecutive Half Hour (1.25 in.)  Storms, Eight Rotations
Treatment
Cw vs C
\J W V W \J
y


DIW vs D y
-L -U


Cz vs D-w
•^

C.y vs Fcz
u. S
v^. _J

E_x vs F_w
s s
.-* j

E,-x vs F..Z
J _)

F..W vs FJ.Z


Storm

A
B
C
A
B

C
A
B
C
A

B
C
A

B
C
- A
B
C
A
B
C
%

6.6
7.9
8.3
5.1
8.4

8.8
3.9
7.0
6.5
N.S.

5.6
6.8
2.1

5.8
7.4
2.1
5.8
7.4
6.5
4.3
8.0
Organic
1
N.S.
5.6X
6.8
3.1
5.6*

6.0*
8.1
8.4
8.8
N.S. ,
/
7.4'
6.5
4.3,
/
8.0'
7.5
N.S.
7.4
6.5
N.S.
7.4
6.5
Matter
Lb/Acre

18.8
21.0
51.1
10.5
22.7

73.7
8.5
22.0
38.8
0.0

6.2
42.5
3.3

45.6
66.5
3.3
45.6
66.5
5.9
37.5
61.5

N.S. ,
6.2/
42.5
0.7
19.4,
/
33.5'
10.5
22.7 ,
73.7'
N.S.

27.7*
51.8
5.9

37.5
61.5
N.S.
27.7
51.8
N.S.
27.7
51.8
Total Phosphorus
% Lb/Acre

.133
.128
.133
.105
.176

.166
.088
.105
.100
N.S.

.095
.108
.055

.128
.132
.055
.128
.132
.062
.149
.142

N.S.
.095
.108
.086
.134*

.145
.105
.176*
.166
N.S. ,
/
.133'
.116
.062

.149
.142
N.S.
.133
.116
N.S.
.133
.116

0.385
0.329
0.783
0.213
0.460

1.413
0.192
0.338
0.626
N.S.

0.098
0.589
0.089

0.817
1.140
0.089
0.817
1.140
0.77
0.739
1.206

N.S.
0.098X
0.589
0.019
0.465

0.849
0.213
0.465
1.413'
N.S.

0.492*
0.909
0.077

0.739
1.206
N.S.
0.492
0.908
N.S.
0.492
0.909

-------
      Table 9  (continued)
t_n
Treatment
C4w vs C4y


D..W vs D..y


C,z vs D..W


C4y vs F5z


E,.x vs F-W


E,-X VS F-Z

F..W vs F-Z
Storm
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
Total Nitrogen
°/° Lb/Acre
.352
.420
.426
.260
.457
.453
.206
.444
-349
N.S.
.320
.363
.331
.384
.381
.331
.384
.381
.197
.369
.346
N.S.,
.320'
.363
N.S.
.312*
.334*
.260
.457
V
.453
N.S.
.354
.317
.197X
.369
.346
N.S.
.354
.317
N.S.
.354
.317
0.99
1.11
2.64
0.53
1.19
3.79
0.45
1.37
2.06
N.S.
0.49
2.03
0.59
2.59
3.37
0.59
2.59
3.37
0.27
1.74
3.37
N.S.,
0.497
2.03
N.S.
1.08,
1.90'
0.53
1.19
3.79
N.S.
1.32
2.54
0.27
1.74
2.83
N.S.
1.32
2.54
N.S.
1.32
2.54
Sand
%
0.00
3.1
1.4
0.00
8.8
5.1
N.S.
13.8
5.3
N.S.
14.4
6.4
N.S.
9.4
4.0
N.S.
9.4
4.0
6.0
7.1
1.4
N.S. ,
14.4',
6.47
N.S.
10.4
8.2
0.00
8.8
5.1
N.S.
5.1
6.4
6.0
7.1
1.4
N.S.
5.1
6.4
N.S.
5.1
6.4*
Silt
%
31.8
34.2
35.9
25.8
30.8
36.7
N.S.
26.7
33.6
N.S.
32.0
34.1
N.S.
37.4
36.3
N.S.
37.6
36.3
23.8
29.0
38.1
N.S.
32.0
34.1
N.S.
29.3
36.2
25.8
30.8
36.7
N.S.
36.1
34.5
23.8
29.0*
38.1
N.S.
36.1
34.5
N.S.
36.1*
36. 5'
Clay
%
68.2
62.7
63.0
74.3
60.5
58.2
N.S.
59.4
61.1
N.S.
54.2
59.4
N.S.
53.1
59.7
N.S.
53.1
59.7
70.7
63.8
61.4
N.S.
54.2
59.4
N.S.
60.2
55.6
74.3
/ ^T * *J
60.5
58.2
N S
tjl • ij *
58.8
•^ *-S • \J
59.1
70 2
/ vy • ^
63 8X
'-' ~J • \J
61.1
N.S.
58 8
-f VJ • (J
59.1
N.S.
58.8
59.1
          No sample

-------
Table 10.   T-Test Comparisons of Treatment Means  for Variables in
           Sediment from each of Three Equal Consecutive Half Hour
           (1.25 in.)  Storms, Three Hay Rotations

Treatment
(1)
H.w vs H.z
4 4
(2)
H,z vs Jw


Storm

A 0
B 6
C 5

A 0
B 6
C 6



.0
.8
.6

.0
.4
.2
Organic
I

0.0
6.4
6.2

2.6
8.5*
7.8
Matter
Lb/Acre

0.0 0.0
10.8 13.9,
15.7 21. 87

0.0 0.826
13.9 25.0
21.8 42.4*


Total
Phosphorus
% Lb/Acre

0
0
0

0
0
0

.0
.113
.112

.0
.089
.079


0.0
0.089*
0.0797


0.059
0.174*
0.156*

q.
0.
0.

0.
0.
0.

0
186
349

0
195
294

0.0
0.195
0.294

0.019,
0.5267
0.879*
Table 10 (continued)
Total Nitrogen
Treatment
(1)
H.w vs H.z
4 4
(2)
H.z vs Jw
Storm

A 0.0
B 0.350
C 0.286

A 0.0
B 0.339
C 0.329
%

0
0
0

0
0
0


.0 0.
.339 0.
.329 0.

.0 0.
.460* 0.
.4537 1.
Lb/A

0 0.0 N.
563 0.639 5
792 1.169 16

0 0.0 , N.
639 1.3587 0.
169 2.387* 8.
Sand


S
•
•

S
7
3
%

. N.
7 0.
4 8.

. N.
3.
2.
Silt


S. N.S
7 34.
3 33.

S. N.S
9 33.
9 30.
%

. N.
6 33
6 30
,
. N.
1 34
7 35


S.
.1
.7

S.
.1
.1
Clay


N.S
59.
50.

N.S
66.
61.
7
/o

. N.S.
6 66.2,
0 61. 0X

. N.S..
2 62. I7
0 62.0
                                116

-------
Table 11.  Enrichment Ratios (E.R.) for Variables in Sediment from Each of Three Equal Consecutive
           Half Hour (1.25 inch) Storms, Continuous Corn.
Treatment Storm
Ax A
B
C
Bw A
B
C
cry A
B
C
C x A
B
C
Ay A
B
C
Aw A
B
C
Az A
B
C
Organic Matter (%)
Soil Sediment E.R.
4.1 NS1
6.9
6.6
3.4 3.8
6.0
6.0
3.7 5.6
6.5
5.6
3.9 5.2
5.7
4.2
4.0 NS
6.8
6.5
4.5 '0.0
8.4
8.2
4.8 0.0
7.6
8.3
_ __
1.7
1.6
1.1
1.8
1.8
1.5
1.8
1.5
1.3
1.5
1.1
	
1.7
1.6
	
1.9
1.8
	
1.6
1.7
Total Phosphorus (7=)
Soil Sediment E.R.
.081 NS
.144
.125
.070 .063
.115
.112
.067 .109
.136
.113
.079 .104
.139
. 23
.077 NS
.142
.121
.083 .000
.146
.141
.093 .000
.145
.158
	
1.8
1.5
0.9
1.6
1.6
1.6
2.0
1.7
1.3
1.8
1.6
	
1.8
1.6
- — -
1.8
1.7
___
1.6
1.7
Total Nitrogen (%)
Soil Sediment E.R.
.218 NS
.368
.362
.190 .200
.306
.310
.196 .288
.344
.319
.189 .250
.360
.329
.196 NS
.348
.329
.229 .000
NS
.405
.238 .000
.377
.419
	
1.7
. 1.7
1.1
1.6
1.6
1.5
1.8
1.6
1.3
. 1.9
1.7
___
1.8
1.7
_ «• •*
_ __
1.8
— ~ _
1.6
1.8

-------
       Table 11.   (Continued)
oo
Treatment Storm
Ax A
B
C
Bw A
B
C
Ccy A
B
C
Ccx A
B
C
Ay A
B
C
Aw A
B
C
Az A
B
C
Sand (%)
Soil Sediment
37.7 NS
6.4
8.6
29.9 3.4
5.8
1.5
35.5 0.5
3.3
6.7
40.5 3.3
5.7
11.0
29.8 NS
2.9
3.2
47.8 NS

6.8
29.6 NS
5.1
4.2
E.R.
— — —
.17
.23
.11
.19
.05
.00
.09
.19
.08
.14
.27
	
.09
.11
___
.28
.14
___
.17
.14
Silt (%)
Soil Sediment
29.0 NS
35.2
31.7
28.5 23.7
38.2
36.4
28.2 29.3
31.9
35.3
27.8 28.5
32.6
36.0
28.6 NS
31.1
31.4
22.4 NS
24.6
29.8
29.0 NS
34.8
32.2
E.R.
. - .
1.2
1.1
.83
1.3
1.3
1.0
1.1
1.3
1.0
1.2
1.3
	
1.1
1.1
_ __
1.1
1.3
___
1.2
1.1
Clay (%)
Soil Sediment
33.3 NS
58.4
59.7
41.6 40.7
56.0
62.1
36.3 70.2
64.8
57.8
31.7 69.7
61.6
53.1
41.6 NS
65.9
65.4
30.0 NS
61.8
54.4
43.4 NS
60.1
63.5
E.R.
___
1.8
1.8
.98
1.3
1.5
1.9
1.8
1.6
2.2
1.9
1.6
	
1.6
1.6
___
2.1
1.8
...
1.4
1.5
               No sample

-------
Table 12.  Enrichment Ratios (E. R.) for Variables in Sediment from each of Three Half Hour  Storms
           (1.25 inch)  8 Rotations.
Treatment Storm
C.w A
4 B
C
c.y A
4 B
C
C z A
B
C
D.W A
1 B
C
D y A
1 B
C
E.x A
B
C
Fsw A
B
C
F z A
5 B
C
Organic Matter (%)
Soil Sediment E. R.
4.3 6.6
7.9
8.3
4.8 NS1
5.6
6.8
4.2 3.9
7.0
6.5
5.2 5.1
8.4
8.8
3.8 3.1
5.6
6.0
4.6 2.1
5.8
7.4
4.6 4.3
8.0
7.5
4.9 NS
7.4
6.5
1.5
1.8
1-.9
_.-
1.2
1.4
0.9
1.7
1.5
1.0
1.6
1.7
0.8
1.5
1.6
0.5
1.3
1.6
0.9
1.7
1.6
_ - -
1.5
1.3
Total Phosphorus (%)
Soil Sediment E. R.
.108 .133
.128
.133
. 080 NS
.095
.108
.063 .088
.105
.100
.111 .105
.176
.166
.091 .086
.134
.145
.104 .055
.128
.132
.084 .062
.149
.142
.079 NS
.133
.116
1.2
1.2
1.2

1.2
1.4
1.4
1.7
1.6
0.9
1.6
1.5
0.9
1.5
1.6
0.5
1.2
1.3
0.7
1.8
1.7
- - -
1.7
1.5
Total Nitrogen (%)
Soil Sediment E. R.
.230 .352
.420
.426
. 244 NS
.320
.363
.197 .206
.444
.349
.264 .260
.457
.453
.201 NS
.312
.334
.260 .331
.384
.381
.233 .197
.369
.346
.250 NS
.354
.317
1.5
1.8
1.9
_._
1.3
1.5
1.1
2.3
1.8
1.0
1.7
1.7
.__
1.6
1.7
1.3
1.5
1.5
0.8
1.6
1.5
_ *. _
1.4
1.3

-------
Table 12 (continued)

Treatment Storm
C.w A
4 B
C
C y A
4 B
C
C,z A
B
C
H-1
° D-W A
B
C
D.y A
1 B
C
Ecx A
5 B
C
F w A
5 B
C
F z A
5 B
C
Sand (7o)
Soil Sediment
30.8 0.0
3.1
1.4
26.4 NS
14.4
6.4
18 . 9 NS
13.8
5.3

20.2 0.0
8.8
5.1
23.8 NS •,
10.4'
8.2 :
17.5 NS
9.4
4.0
24.5 6.0
7.1
1.4" -
27.3 NS
5.1
6.4

E. R.
_ m. —
.10
.05
___
.55
.24
_ _ .
.73
.28

	
.43
.25
— - -
.44
.34
_._
.38
.16
.24
.29
.06
__-
.19
.23
Silt (7=)
Soil Sediment
23.4 31.8
34.2
35.9
27.8 NS
32.0
34.1
27.2 NS
26.7
33.6

28.5 25.8
30.8
36.7
28.9 NS
29.3
36.2
29.7 NS
37.4
36.3
28.8 23.8
29.0
38.1
29.1 NS
36.1
34.5

E. R.
1.4
1.5
1.5
___
1.2
1.2
	
.98
1.2

.91
1.1
1.3
......
1.0
1.3
M •» «
r.3
1.2
.83
1.0
1.3
- — -
1.2
1.2
Clay (7o)
Soil Sediment
45.8 68.2
62.7
63.0
45.8 NS
54.2
59.4
53.8 NS
59.4
61.1

51.3 74.3
60.5
58.2
47 . 3 NS
60.2
55.6
52.8 NS
53.1
59.7
46.7 70.2
63.8
61.1
43.6 NS
58.8
59.1

E. R.
1.5
1.4
1.4
	
1.2
1.3
	
1.1
1.1

1.5
1.2
1.1
.. _ -
1.3
1.2
« «• —
1.1
1.3
1.5
1.4
1.3
_ — —
1.3
1.4
No sample.

-------
Table 13.  Enrichment Ratios (E.R.)  for Variables in Sediment from each of Three Equal Half Hour
           Storms (1.25 in.), Three  Hay Rotations
Treatment
(1)
H,w


(2)
v
•-^f

(3)
Jw


Storm

A
B
C
•.
A
B
C

A
B
C
Organic Matter
Soil Sed. E

4.5 0.0
6.8
5.6

4.0 0.0
6.4
6.2

4.9 2.6
8.5
7.8
.R.

. 	
1.5
1.2

	
1.6
1.5

0.5
1.7
1.6
Total Phosphorus (%)
Soil Sed. E.R.

.045 .000
.113
.112

.089 .000
.089
.079

.100 .059
.174
.156

. —
2.5
2.5

	
i.o
0.9

0.6
1.7
1.6
Total Nigrogen (%)
Soil Sed. E.R.

.222 .000
.350
.286

.207 .000
.340
.329

.288 .000
.460
.453

	
1.6
1.3

	
1.6
1.6

	
1.6
1.6

-------
          Table  13 (continued)
to
N>

Treatment
(1)
H.w
A
*T

(2)
Hz
hf.

(3)
Jw



Storm

A.

B
C

A
B
C

A
B
C
Sand (%)
Soil Sed.

16.7 N.S.1

5.7
16.4

34.5 N.S.
0.7
8.3

24.1 N.S.
3.9
2.9

E.R.

___

.34
.98

	
.02
.24

	
.16
.12
Silt (%)
Soil Sed. E.R.

28.4 N.S.

34.6 1.2
33.6 1.2

27.0 N.S. 	
33.1 1.2
30.7 1.1

28.9 N.S. 	
34.1 1.2
35.1 1.2
Clay (%)
Soil Sed.

54.9 N.S.

59.6
50.0

40.5 N.S.
66.2
61.0

47.0 N.S.
62.1
62.0

E.R.

— _ —

1.1
.91

	
1.6
1.5

	
1.3
1.3
No sample

-------
N>
OJ
       Table 14.  Analysis of Variance of Several Crop Rotations .

                  (1.25 in.) Storms
Three Consecutive Half Hour

Rotation
Aw - Bw
C w - C y
y y
DjW - Djy
H.w - H.z
4 4
Treatment
Manure (6 T/A)
No Manure
Storm
A
B
C

Gal. /Acre

9678a1
8960a
11202a
5926b

8664a
9219a

725a
8080b
18020c
Runoff

Solu.
Ortho.
NO.-N (PO.)-P
(Lbs //Acre) (Lbs . /Acre

0.48ab
1.04b
1.05b
0.22a

0.65a
0.74a

0.04
0.57
1.47

O.OOlSa
0.0012a
0.0067b
O.OOOSa

0.0033a
0.0020a

0.0002a
0.0037b
0.0039b

Soil
Organic
Matter
Tons/Acre (Lbs. /Acre)

0.146a
O.lSla
0.172a
0.087a

0.140a
0.139a

0.019a
0.173ab
0.275b

22.09a
19.95a
20.08a
10.37b

18.10a
18.14a

15.20a
18.79a
31.03b
Loss

Total Total
Nitrogen Phosphorus
(Lbs. /Acre) (Lbs. /Acre)

1.02ab
1.21ab
1.42a
0.53b

1.17a
0.92a

0.23a
0.88a
2.02b

0.404ab
0.364ab
0.569a
0.171b

0.418a
0.335a

0.092a
0.347ab
0.693b
            Means followed by the same letter are not  significantly different at  5% level.

-------
    15.  Analysis  of Variance  of  Two Continuous Corn Rotations from Three Consecutive  (1.25  inch)
        Storms.
Factor
Level

Gal. /Acre
Rotation
Aw - Bw
Ccx - Ccy
Treatment


Storm



30# P205
90# P205
Manure
(6T/A)
No Manure
A
B
C

9678 a
13030 b
8772 a

13936 a
1718 a
11347 b
20997 c


Mean
Runoff
Soluble
N03-N Ortho-(PO,)-P
(Lbs./Acre (Lbs./Affre)

10.48 a
0.73 a
0.79 a

0.42 a
0.05 a
0.51 a
1.25 b

0.
0.
0.

0.
0.
0.
, o.

0018 a
0067 b
0041 a

0045 a
0005 a
0047 b
0077 c


Tons/Acre

0.
0.
0.

0.
0.
0.
0.

146 a
260 b
143 a

263 b
060 a
197 b
351 c

Soil Loss
Organic
Matter
(Lbs ./Acre)

22.1 a
32.3 b
20.3 a

34.1 b
9.21 a
29.97 b
42.41 b


Total
Phosphorus
(Lbs./Acre)

0.404
0.684
0.416

0.672
0.179
0.605
0.849

a
b
a

b
a
b
b

Total
Nitrogen
(Lbs./Acre

1.02 a
1.79 b
1.07 a

1.74 b
0.42 a
1.36 a
2.43 b
Means followed by the same letter are not significantly different at 5% level.

-------
      Table 16.   Correlation and Regression Coefficients of Surface Runoff (Gal/Acre) p.nd Soil Loss  (Tons/Acre)
                 During Three Storm Frequencies (Years).
                        Correlation Coeff.
Regression Coeff.
Storm Frequency
Crop
Continuous Corn
Mixed Rotations
Continuous Hay
2
.84
.91
.41
10
.63
.51
.88
20
.54
.50
.86

Y =
Y =
Y =
2
-,.Q14+4xlO~5(X)
-.003+3xlO~5(X)
.OD4+lxlO~5(X)

Y =
Y =
Y =
10
.087+1x10
.063+1x10
. 004+2x10

~5(x)
~5(x)
~5(x)

Y =
Y =
Y =
20
.187+lxlO"5(X)
.144+lxlO~5(X)
.012+2xlO"5(X)
N>
Ul

-------
Table 17.  Correlation of  Selected  Constituents  in the Soil and
           Sediment Derived from the  Soil  and  Enrichment Ratios
           (E.R.) from Each of  Three  Consecutive Half  Hour (1.25 in)
           Storms.
A
Constituent
Organic matter
Total P
Total N
Clay
Silt
Sand
r
.81
.60
.81
-.32
-.13
-.41
E.R.
1.8
1.8
1.8
1.6
1.0
.08
Storm
B
r
.70
.64
.77
.01
.40
.07
E.R.
1.6
1.6
1.7
1.4
1,2
.25
C
r
^54
.65
.78
.07
-.03
.01
E.R.
1.5
1.5
1.6
1.4
1.2
.21
Table 18.  Correlations of Percent Solids in Runoff and  Selected
           Variables for Each of Three Consecutive Half  Hour  (1.25  in)
           Storms.


                                        Correlation Coeff.
Storm
A
B
C
Mean
% Solids
.44
.52
.41
Std. Dev,
of mean
.38
.31
.21
t
Runoff
.16
-.17
-.07
% Slope
.17
.10
.12
Aggregate
% O.M. Stability
.03
.26
.03
.008
.24
.06
% clay
-.02
.16
.07
                               126

-------
Table 19.  Correlations .Among Nitrate Nitrogen and Soluble Ortho-
           Phosphate  (PO^)-P in  the Soil and Their Concentrations
           in ^unoff Water for Each of Three Consecutive Half Hour
           (1.25 in) Storms.
             Concentration  of NO  -N  in Runoff
A
NO -N in Soil .21
Storm A
Storm B
B
.01
.28

C
.28
.21
.59
             Concentration of Solu.  PO.-P in ^unoff
A
Solu. PO.P in Soil .15
4
Storm A
Storm B
B
.14
.71

C
.32
.55
.83
                                127

-------
Table 20.  Content and Correlations Among Constituents in the Soil.
     Constituent        Mean   Std. Dev.       Correlation Coeff.
                         %     of mean     P     N    Cl    Si      Sa
Organic Matter (O.M.)
Total P (P)
Total N (N)
Clay (Cl)
Silt (Si)
Sand (Sa)
4.30
0.084
.22
44.0
28.0
29.0
1.04
0.030
0.06
10.0
4.0
12.0
.66 .91 .24 .08
.75 .12 -.03
.22 .11
.17


-.23
-.09
-.23
-.93
-.49

Table 21.  Content and Correlations Among Constituents in the Sediment.

     Constituent       Mean    Std. Dev.       Correlation Coeff.
                        %      of Mean    P     N    Cl    Si     Sa

Organic Matter (O.M.)   6.9     1.9      .69   .81   .20  -.10   -.24

Total P (P)              .13     .038          .75   .12   .03   -.20

Total N (N)              .37     .089                .19  -.06   -.26

Clay (Cl)              61       8.6                       -.46   -.67

Silt (Si)              33       5.8                              -.23

Sand (Sa)               5.9     6.9
                               128

-------
Table 22.  Stepwise Regression of  Selected Constitutents on Other
           Constituents  in  Sediment.
                                                           Multiple
            Stepwise Regression Equation               Correlation Coeff-

% P = 0.011 + 0.327 (% N)                                       .75

% P = 0.009 + 0.249 (% N)  + 0.0045 (% O.M.)                     .77

% N = 0.108 + 0.0378 (% O.M.)                                   .81

% N = 0.076 + 0.0257 (% O.M.)  + 0.876 (% P)                     .86

% N = 0.086 + 0.025 (% O.M.) + 0.866 (%'p)  -0.00084 (% Sa)      -86
                                 129

-------
                      Analytical Determination


Surface Runoff

The sample from the Coshocton wheel was transported to the laboratory
and weighed on a solution balance.  While mixing vigorously to keep
the sediment well mixed, three 40 ml (approximately) samples were
placed in tared 50 ml centrifuge tubes.  The total weight of the sample
plus tube was recorded and the samples were centrifuged at 17000 rpm
for 30 minutes in a Sorvall superspeed centrifuge.  The supernatant
solutions were poured off and combined into one solution sample.  The
residue in the tube was dried and the tube plus dry residue was re-
weighed.  From these data, the sediment content was calculated.

The supernatant solutions were analysed for NH^-N, NO^-N, soluble P and
total P according to procedures listed under Analytical Procedures.

The above procedures were carried out the day following collection of
the sample in most cases.  If this was not feasible, they were stored
in a refrigerator until analysed.

Bulk samples of solids were prepared as follows.  Five ml of a solution
of CaCl2 which contained 3 g of CaC^'ZlLgO per 5 ml was added to each
gallon of the runoff sample.  After 48 hours the supernatant solution
was siphoned off and the remaining slurry kept frozen until it was
convenient to process it further.  After thawing, the slurry was
filtered using a Biichner funnel and suction.  The samples were oven
dried and ground.

Tile Drain Effluents

These samples were usually clear and so no separation of solids was
carried out.  The solutions were analysed for NH,-N, NO--N, soluble P
and total P according to procedures listed under Analytical Procedures.

                        Analytical Procedures

Solutions

The determinations were made with the Technicon Autoanalyser systems
using the following procedures.

Ammonium.  This procedure involves the reaction of ammonium, phenol,
and hypochlorite in an alkaline medium which yields a blue color.
Nitroprusside is added to increase the sensitivity of the determination.
The structure of the blue-colored compound is believed to be closely
related to that of indophenol.  Reference:  Russell, J.  Biol. Chem.
156, 457, (1944).
                              130

-------
Nitrate plus Nitrite.  Nitrate  is reduced to nitrite by copper and
hydrazine sulfate in alkaline solution.  The nitrite is determined by
diazotization of sulfanilamide  in phosphoric acid and coupling with
N-(l-Naphthyl) ethylenediamide  dihydrochloride.  The resulting azo-dye
is pink in color, and has an intensity related to nitrite concentration.
Reference:  Jacobs and Hdchheiser.  Anal. Chem. _30_, 426, (1958).

Soluble Orthophosphate.  Ammonium molybdate in acid solution reacts
with orthophosphate to form a heteropoly acid (molybdophosphoric acid).
This acid is reduced by stannous chloride and hydrazine sulfate to
form the intensely colored molybdenum blue complex.
Reference:  Fiske and Subbarow.  J. Biol. Chem. 6±, 375, (1925).

Total Soluble Phosphate.  Polyphosphates are hydrolyzed and organic
phosphates oxidized by heating  with potassium persulfate solution.
The resulting orthophosphate is determined by the procedure described
under soluble orthophosphate.
Reference:  Menzel and Corwin.  Limnology and Oceanography, 10, 280,
(1965).

Particulate Matter

Total Phosphorus.  The soil is  ashed with magnesium nitrate at 55°C.
The residue is heated with hydrochloric and nitric acid to hydrolyze
polyphosphates to orthophosphate and to precipitate silica.  Ortho-
phosphate in the resulting solution is determined by reacting with
molybdic and vanadic acids to form the yellow heteropoly molybdovanado-
phosphoric acid.
Reference:  Methods of Analysis for the Association of Official Agri-
cultural Chemists.  6th Ed., A.O.A.C., Washington, 1945.  Kitson and
Mellon.  Anal. Chem. 16, 379, (1944).

Total Nitrogen (Kjeldahl).  Soil is digested with sulfuric acid,
potassium sulfate and copper sulfate to convert nitrogen to ammonia.
The resulting solution is made  basic and the ammonia is distilled into
boric acid.  The ammonia is then titrated with standardized sulfuric
acid.
Reference:  Methods of Soil Analysis.  Am. Soc. Agron., Madison,
Wisconsin, (1965).

Organic Matter.  Soil is treated with a solution of potassium
dichromate and then concentrated sulfuric acid is added.  The heat of
dilution of the acid is utilized to bring about oxidation of the soil
organic matter.  After cooling  the excess dichromate is measured by
titration with ferrous sulfate.  The amount of dichromate reduced is
related to organic matter concentration of the soil.
Reference:  Greweling and Peech.  "Chemical Soil Tests."  Cornell Univ.
Agr. Exp. Sta. Bull. 960 revised.  (1965).

Particle Size Analysis.  Soil is treated with hydrogen peroxide to
destroy organic matter and then washed and centrifuged to remove
                               131

-------
dissolved mineral matter.  The soil is dispersed with sodium hexameta-
phosphate and particle size distribution is measured by the Bouyoucos
hydrometer method.
Reference:  Bouyoucos, Science 64,  362, (1926).
                              132

-------
                             APPENDIX B

Table 1 is a presentation of the  treatment schedule on the field scale
water quality study.  The beans/wheat plots actually receive almost
twice as much fertilizer as does  the corn because of the fall ferti-
lization of wheat.  The wheat already established from the previous
fall received little or no fertilizer during the next hydrologic year.

Tables 2, 3, and 4 are computer programs written in Fortran IV and
processed on the IBM 360/65 computer.  Table 2 is a calculation of
surface flow in gallons/acre or acre inches of water.  In addition, the
centroid time, time of maximum discharge and discharge for any given
time interval is calculated.  Table 3 is similar except it is used to
calculate tile discharges in gallons/plot for any given time interval
on a monthly basis.  Table 4 is a program to summarize by plots the
flow and the quantities of nitrogen and phosphorus lost from the land
surface.

Table 5 is a nutrient balance for nitrogen (N) and phosphorus (P) inputs
and outputs on the field scale water quality study.  Fertilizer inputs
were calculated on actual amounts of mineral fertilizer added between
4/1/70 and 3/31/71.  The crop residue return was calculated on 49% of
the stover weights of corn and wheat to equal grain yield whereas the
bean vine weights equal the weight of beans.  The alfalfa yield (grown
in succession with good managed wheat) was 2,000 pounds/acre.  These
yields were multiplied by the appropriate percentages of N and P to
arrive at a return value for crop residues.  It was assumed that 80%
of the residue decomposed in the  first year and the remainder was tied
up in soil organic matter.  Mineralization inputs of N were derived by
using 58% of the total weight of  organic matter as organic carbon.
With a C:N ratio of 14:1, 1/14 of this weight is N.  The product of
this is multiplied by a constant  of 1.2%.  Mineralization of P is
assumed to be 14 pounds/acre/year based on unpublished data of S. Reid
(Cornell University).

Total removal of N and P by the crop was developed by using the product
of the yield and the percentage of N and P contained in the crop.  Run-
off, sediment and deep seepage losses of N and P are the product of
flow or soil loss and the concentration of the nutrient in question.
The total loss to the environment is the total of the losses due to
surface runoff, sediment and deep seepage.  Denitrification and
fixation of phosphorus was determined by subtraction.

Table 6 presents the total accumulative runoff and nutrient losses for
the 1970-71 hydrologic year.  The values are broken down into the
treatment variables studied.  The Least Significant Difference (L.S.D.)
was calculated to determine real  differences between variables.
                              133

-------
Table 1.  Fertilizer Application Rates on Corn, Beans and

          Wheat Plots.  4/70-3/71
Element

Fertility
  Level
     N
High   Normal
               P205
           High   Normal
                             K20


                        High    Normal
Corn

Beans
Wheat

Wheat
 275

 275
 215

  0
15

15
75

40
(Pounds Per Acre)

     150     30
     150
     130

      0
30
50

20
100     100

100     100


100     100
                               134

-------
Table 2.  Surface Flow Program.
Variable Name

   AINT1

   AINT2

   AMON

   AREA
   ARM
   CENTR
   CONCT

   CPDATE
   DEPTH
   DISCH
   HMIN
   HOURS

   ICHECK

   ICLASS

   ICODER

   IDAY

   IMON

   IPLOT

   ISTOR

   IYEAR

   KLASS
   KODER
   NCHECK
   NCLASS
   NCODER
   NDAY
   NMON
   NPLOT
   NSTOR
   NYEAR
   PROG
   QMAX

   STAGE
   TARE
                 Description

Outer limit of a time interval needed to
  calculate TARE
Inner limit of a time interval needed to
  calculate TARE
Accumulation of the product of time and dis-
  charge
Total gallons of runoff for a given storm
Time accumulation
Centroid time of storm
Time lag between beginning of runoff and the
  centroid of the storm
Date of computation
Acre inches of runoff
Discharge in gallons for 12 in. H-flume
Time point (minutes at start of storm)
Time point (hours at start of storm -
  military time)
Control to reset NCHECK number if different
  from previous record
Control to reset NCLASS number if different
  from previous record
Control to reset NCODER number if different
  from previous record
Control to reset NDAY number if different
  from previous record
Control to reset NMON number if different
  from previous record
Control to reset NPLOT number if different
  from previous record
Control to reset NSTOR number if different
  from previous record
Control to reset NYEAR number if different
  from previous record
Evaluation of KODER
Name of data transformer
Number code of checker
Number code of KLASS
Number code of KODER
Day
Month
Plot number
Storm number
Year
Person compiling data
Maximum discharge that occurred during a
  storm
Stage height (to 0.1 feet)
Product of the time interval and the dis-
  charge
                               135

-------
Table 2.  (Continued)

Variable Name                               Description

   TIME                   Time point (hours and tenths of hours)
   TQMAX                  Time of maximum discharge
                              136

-------
Table  2.  (Continued)
                 .A."! mn\>M|-H-AIIKIHfl  12
                 iiv  i l i->h <> on ), HI iu«s <><)<», HI., !*!<;«,<» ,oisCH(?oo) , STAt;t(2oo)
              * w   Kl M.itrK (M ) , K'. aSS (4 )
               IIIN  AHIIUKSI 11 ) , Ai-lIM 11 ) ,/sSTAGhl 11 )
                J=l.2od
        HwIlM( ,| )=().(.)
        STAUt ( J)=0.0
        i)i SCH( ,i )=o.o
        T I MtU 1=0.0
        M=l
        HtAI)  HO, (Klll)t«( J) ,J = 1 ,9)
    HO  FORNifiT (9AH )
        HtAI)  HI , (Kl.fiSS( .1) ,J=1,9)
    HI  KI«MAT(yAH )
        READ  MM), p«or, .CWATt
   4OO  HIIKMAT  ( 2AH ) /
    (Sti  Kt Al)  6(),Npi. ()T ,NMIlN,N|)AY,NYt AH , NCI. A S S , NCOOfiR , NCHtCK ,NST()R ,
       1 ( HIHIKSJ J ) .HrtlNl .1 ) ,STAf,t (J ) ,.I = M,'. )
    60
    6H  IH (NP'.IIT-IPI.OT )fth,M ,
    61  I F( NSTIIK-I STIlk )6h,62,
    62  1 Pi_nT = NPi_()T
        II)AY = NI)AY
        ICHtCK=NCHtCK
        1C". ASS = NO. ASS
        DO  S*l J = M,i.
        I F( HDUKSI J ) )?()(), 2 00,? 02
   200 1MHMIM J  M20l.201,202
   201 Ih( STAGt(J) 141,41,202
              J )=
    91  ClllMTINUt
        M=l. + 1
        (ill  TO 6b
    66 on  139  J=l,ll
        AHOUKSI J)=0.0
        AMlN( J)=0.0
   139 ASTAGtt J)=0.0
        1)11  14O  J=l,ll
        AHf)tlKS( J)=H()IIKS(K)
        AMIN1J )=HMlf4(K )
        ASTA(,h(J)=STAGt(K)
   140 CflNTlNDt
        1)11  IO1  .1=1 «N
        1H STA(;h( J ) )42,7,H
    42 »>HIivT  17,HI)l)HS( .» ,HMJN( J) ,STAf,t( J)
    17
      7
                                        137

-------
Table  2.  (Continued)
         t;:i  Hi  101
       «  I t- ( SI /VI-, i- ( .1 )-o. 04 4)4.4, ) u
       4  n I v. H (  ,| ) = i  >>o . o* | ST Ai-r ( ,1 ) ** I .'/
         (;n  Mi  lul
      In  TH ST/M-I- ( ,1 ) -('. inn) 1 i . 1 1 , 1 >
      11  MI SC. n(  ,| ) = ?f>> .<)* ( STrtM- ( J ) *f I . i
         (in  Tii  101
      i/  IH sT/n;t u >-o.;>?;> i n. M. 14
         (•;n  ni  10]
      14 IF  ( SF AGt ( ,1 )-i).Y(io) ll. 11 , 1 ^>
      Ih HRINT  '/> \ (). HIIIIK S( ,1 ) ,HM I N ( J ) . ST A(ih ( .1 )
    >-lo hMKMArH STAdb  nVh-^  wflTli\'(>  Cn«vt  . ^t- 10. /?
      1^ 0] SCM( J ) = 7().H.()v( STA(;r( J
    101 CIlNTlNlIt:
         A MM = 0.1 1
         A«tA=0.0
         C'lNC T = 0.0
         1)1)  10?  J= 1 .N
         I F( UHAX-DI SCN( J ) ) 1 H, 10? . 1 0V
      1 H (.IM A X = I ) ] SC H t .1 )
         TUMfiX = T I Ml- ( J ) /Ml. (i
     l(>/» CUNT INI it
         '. =M-1
         no  103  J=?,l
         A]MT1 = T iMh ( .) )-T I Mt ( .1-1 )
         IK ( AINT1 )30,H1 ,31
      30 a INT 1 = A I wr 1 + 1 44o.o
      31 AINT? = TIi«M.I + l )-T Mh (.1)
         IH ( A INT? ) 3?; 3^.3 "4
      3? A |NT? = A INTJ- + 1440.I)
      33 TAHt=( AINTH-AINT^ )*«.'. ^>*DI SC^J J J
         AWM = AWM-t-A II\lT 1
         A«t- A = AK t A+T AK t
     103 Oi IN 1' INI it
         OH  13 TI'«I- ( .1 )=T IMC< J
         Ct VTK = />'••! n« / ( A«t-A*hO.(i)+T I'v't- I 1 )
         T H ( CtNT«-?t>.0<> 140,41 ,41
      41 OeNTK=r.tNTK-?4.00
      40 PRIiMl  70
      70 KUKMA'I ( « 1 ' ./ .4«JH KOT    S1HHM    MIIMVH DAY YtAM
                DATA  KATIMfi    MAX  IUSCH 1 | Mh    C H xj \ k( 1 1 I )  flilt-     TIlTAI.  GALS
               )
        SlJKKACtr  A^itA 'II-  H'.IITS  I NO. Hi) Iwf,  HITCH IS  0.9  AO«hS.
         CHNOT=Ct'MTK-T I MM 1 )
          IF (01 INC T 1403 1 404, 404
     4O3
     404
         J=ICMI)tK
         K=ICH|-r.K
         I t- ( K )44 ,14.
                                        138

-------
Table 2.  (Continued)
     Hi '. = 1C1. ASS
         1 H '. ) H is , *>fc , r> 7
     i ^ '. = 6
     S7  P* I vT  71 , ] Ml. 1 1 r . I S FUR , I AH n\i, | II/>Y , 1 YH;, •< . K> iiir-w ( j ) . K, im-* ( * ) . KI. A <; S ( I. I .
        1 Ui»i ft X , I (.11*1 A X , C h iv F M , A K I- A
      ? I  M I* PI ft T     (14, ?! H, I*-,, J4, /x, ft M.IX. AM. -,*, « «.t-i>. >•,(-!-<.:•(, MO. ?, lux,
        r> ici.? i
         p»< INT 401
                 ( / /t> i H naifi AsstMHi.hw    r.iifiKii TA r |ii'»i  HATJ-    iiv
                          t-F  in  r. tMTNMin  )
                          ,r.^i)ATb .HhHTn,  r.imr.T
         hi IX W ATI 4X, AH , 1 OX  , AH. 1 UX ,f- I .H. 1 l)X,(- 1 U.'-J// I
            I N T  74
                  // ^7H        TlMh             STAO  niSCH4«f,t-  .
                         TlMh           STA(,I-  ill
         M=l. -t-1
         no 116 J=I.M
         ! = ,)+'.
         M« 1 'NT  7 5, "Mil* S( J ) . HMlN( .1 ) , 1 !'•'•»- ( .1 ) . ST A(,r ( ,1 ) ,1)1 SCH( ,1 ) , H(Hl«S ( I )
        IHM]IV( I ) ,T Ti"t ( I t ,STA<;- ( I ) .n| SCH( | |
          DM  1?? J=1.200
          Hinmsf j ) =().(>
          HI'. I IV ( ,1 ) = O.I I
          STAGb ( .1 )=0.0
          D! Sf.H( .) ) =0.(l
     1 ??  r IMC ( .1 )=0.0
          I M NSTu«-4HS ) 1 /• 1 .
     121  M = ()
          ,->=!
          '. = 11
          on  i4i j=i . 1 1
                         '
          HM 1 iv ( ,1 ) = Aw I i\i { ,1 )
          sTAi;t ( .1 ) =Asr Ai,h
     14]  r.m'
                                           139

-------
Table 3.  Tile Flow Program
Variable Name

   AVE

   CP DATE
   DAYTOT
   DEPTH
   ICHECK

   ICLASS

   IDAY

   ILA
   IMON

   IYEAR

   IPLOT

   JA
   JOHN

   KLASS
   KODER
   MARY
   N
   NCHECK
   NCLASS
   NCODER
   ND
   NDAY
   NMON
   NPLOT
   NYEAR
   PROG
   Q
   S
   ss
   T
   TT
                  Description

Average stage height for each day of the
  month
Date of computation
Accumulation of discharge for monthly total
Acre inches
Control to reset NCHECK number if different
  from previous record
Control to reset NCLASS number if different
  from previous record
Control to reset NDAY number if different
  from previous record
Date of the following day
Control to reset NMON number if different
  from previous record
Control to reset NYEAR number if different
  from previous record
Control to reset NPLOT number if different
  from previous record
Number of days in a month
Date + days past initiation of flow (current
  date)
Evaluation of KODER
Name of data transformer
One less than the number of days in a month
Number of data points
Number code of checker
Number code of KLASS
Number code of KODER
Days since start of discharge
Day of month
Month
Plot number
Year
Person compiling data
Calculation of flow in g.p.m. for the current
  date and counter for the initial data on
  a record
Entering SS into a T and S matrix
Discharge point (stage height - to 0.01 in.)
Entering TT into a T and S matrix
Time point (hours, min. - military time)
                              140

-------
Table  3.   (Continued)
          HKMUKA,.'.  T P. t

          k h/s'_ v,<     K I H lh'< 14 ) , .()
          DM  100  .1= I .All
          DM  100  K=l,Ao
          T ( .) ,K ) =().()
          S ( .1 .K ) =0.0
     1 OO  i.i< ,1 ,K ) =O.O
          DM  !(>'•»  J= ) , AO
     1 O H  N ( J ) = O
     ? OO  P t A I) A , ivpi_ i IT , '\'M i H , «iYt- A •<»*'".',. A SS . '• T. I iO»- * , ™r Hf-r. K . .vo A Y . T T ( 1 ) .SSI 1 ) ,
         1 I M|)( J ) , IT ( .1 I . SS  ( .1 ) . ,) = ? . 1 O)
            ' IR "i a 1 ! '-i I ? , H 1 1 .  1 ? . i- '•<. 1 . t- -t.  ', 4 ( n . i-  '< . I . f- ^. > I )

            F ( KI K>I_ i ] T -4 s ) 1 11, 1 / o , 1 ? o
            (- ( I P'j MT-'VHl. Ml ) 1 >0 .HOI . 1 ?0
*iOOO
   Id
  K(l 1
            t- (  I Nil IPM-IM./I I|M ) ) / O . H O /• , \ ?!}
            iVM Ihtr -\}Kf MM
            Ytr flk=MYtr Aw
                   = ivr.i_ ASS
           1C "Hf. K = IVC HhC K
           I D A Y = iV 11A Y
          NI IM K h * = 1
           P. A = IM ( 1 i) A Y ) + 1
           IF(AO—P.A)?li,??.??
      ?.0  PKlNf  ?.l,lDAY»lN|i'M«I
      ?1  HHKMAT ( //'MuMi  pnl'N'TS  TMII  '. «fci(,r  t-H*  S TMR Af;tr . A If.//
          f;n  Ti1  /:'»<)
      ??  T( 1 DAY, |i. A )=TT ( 1 )
          S( I DAY, p. A ) = SS{ 1 )
          u ( IDAY, p. A ) = A . o H 4 ?_ * ( S( IDAY. p. A
          N( I.IAY ) = p. A
           IF( T( I DAY, p. A ) C>,(S.^
       •>  PRINT  ?.3.T ( IDAY , P.A ) ,S( ID»Y,  P.
      ?b  FMKiiAT ( ////H  AftSMHD  TI"1:  '''«  STAM-  ,?MO.?,6H  DATt  ,
       6   IF(S( IDAY, P. t ) ) Y.H.H
       7  PRINT  ?t>, T( IDAY, P. A ) ,S( IHAY,  P. A ), IDAY. JMIN, I
       H   Ih(S( IDAY , P. A )-A.?b ) HOO,H(IO,4
       4  PRINT  ?.V,T( IDAY. P.A ),S( I'!AY,  P.A), IDAY,I«"MN,I

     HUO  CMwTlMMh
          1)1-  llll   .) = ''' , 1"
           IK S S ( ,M ) 4 y •*. , 4 ? ^ . 4 14
                                            141

-------
Table  3.    (Continued)
     w?i  1 1- ( TT{ .1 ) )  1 <>] , MM ,qj q
     4)q  jiiHi\i= II)AV + IVI)( j )
           1 1. ft=iM( Jl IH-vl ) + 1
           1 |- ( 4(1- 1 1. A  )>• X , >'«  . ?4
      ?'+  iJUjlMT  ;/ 1 ,  I MAY« I'O'fi'M, I YhAK. 1 P'.HT
           r,n  TI  ?nn
      ?4  1 ( JIIHISI, IL  A ) = )T (J )
           S( JilHts', I1.  A ) = SS( .1 »
           '-'( JIIHM , ] I.  ft )=4. (1~*>-)'/* ( S( .I1 'HIV , I'. l\
           IM ( Jl IH V ) = H ( ,11 IHN ) + 1
           TM T( JIIHM, ii. A  ) )  i ) , i?. 1 y
      11  HK INT  /b ,  T ( JIIHV, 1 1. ft ) ,S( ,1  "-M, p. A )  . ,|l|w-i. | WIN. I
      \S  ! F ( S( JIIHM, 1 1. A  ) )  I X. 14, 1 U
      13  P« I MT  ?*> .  T ( JHMV, ] . a | ,\( >l'l-"l. !'. A )  , JMH.V. I i.ld si , I
      14  1 F I S( JIIHIV, IL A  )-<•.;>'> ) 1 Hi , 1 Hi , I h
      1ft  P« IN r  ?7 ,  T I JIIHiS-i  , p. A ) , S ( .In" v, 1 1. A )  . .11 'H.  , I M( I'M. 1 YcAK
     1O1  C i IN T INI It:
           (ill  Til  ?()!>
     120  Ml  Tl)    ( 1,?. 1  .'-S, 1 ,"-), 1 , 1 . -S, I ,'4, I )  . l-i'N
        1  JA = M
           fill  IH  140
        ?.  I F (! Yt-AK-ft4 ) HOX  .MH, «()'•(
           I (- ( lYhrAK-  ftw )   H(IH
           JA=?8
           f4(l  Til  14O
           JA=?9
           (;il  Til  14O
           J A = 3 ( I
           MAWY=JA-1
           DM  in1?  J=l ,HM
           I =M( J )
           I (- ( l-l ) H)*>.'>? .'i?
      b?  ClIiSjT iNllh
           Di|  £ -4  K = 1 , 1
           '. = l + l-K
           1^ (T ( J .1. )-?4 .0)
      4)  T ( J.I. ) = T ( J.L )-^
           H=M( J+l )
       4V  C n M T I isi 1 1 h
           1)11 41
           i. I. =w+?— i
           S( J+l ,1.1. ) = S( J + l .1.1. -1 )
           T ( J+l ,1.'. )=T (J+l ,i.'.-l )
           (-!( J+l .I.L )='•!( J+l .'.'. -1 I
       4h  CHisiT JivllH
       4*  1 H T ( J+l ,? ) )44 ,4S ,4fc
       4^  T ( J+l .•/ ) = T ( J + l , 1 )
           S( J+l ,/ )=S( J+l , U
           U( J+l ,? )=U( J + l , V )
       4ft  T( J+l , 1 ) = T( J,i. )
           S( J+i . 1 ) = S( J,'. )
           i-M J + l , 1 )=U( J,'. )
           T( J,'. )=().()
           S( J,1. )=<>.<>
           i) I J .'. ) =n.o
           IV ( J+l )=M( J+l ) + l
           l«( J )=H( J ) -1
       4'-(  Cm" T !'vl)h
      i n^  r.iiHiT INI ih
                                             142

-------
Table  3.   (Continued)
          mi  ?o6  .1=1,40
          I=w07,?<»
          I (-( r ( J . 1)
          c i) N r i i\i u t
          nil  2:m  K=] ,
          n j.i. > = T (j.i.-i j
                I=S( J.i.-1 )
          '•'< J.1.
          T( J.l
          S( J.I
          '•' ( J , 1
= 0.0
                  = 0.0
     >06
          IO
          ('()  TO  /()0
     64v ina=i+i
     yoo r.Mi\T
          1 1- 1
          '.HI. n = j + 1
          ? K( IM( J ) 11 HA, 1 H6. 701
     701  1 H SP. II'. U. 1 ) ) 1 (i*. 70^
     70^  K=N(J|
          DM  60  '. ='. II'. ll.^'s
          1 H NC. (-1 )6D,6l ,64^
     *4-l  IF ( S(i. , 1 ) )60,6? .61
      6o  CIIN r pMiit
      61  M A = 1
          <;M  TII  63
      6?  MA=?
      6;-i  T A = T ( J.K )«-J*?<..(»
          SA=S( J.K )
          r H = T 1 1. . t" A i +i. * ? A . 1 1
          SH=S(I. ,MA )
          A= fH-Tfl
          IF ( A )6 f ,6*>,6'»4
     ft^4  I HI SA ) 106,6S.6i
      6W  I F ( SM ) 106 , fl ,64
      64  S'.I)MC= ( SH-SA ) /A
          DM  MOO JACKS'. ML M.1.
          si JACK, i I = SA + SL MKt*(
          IF (|M( JAC<
     MS**  M(.IACK) = 1
     400  (->( JACK. 1 )=4.()^y?w( S( JACK, ))**/.
      fl  !='.-!
          GM  TM  64V
      6^  MKINT  66, J, I ( J,K ) , IMMN. I
      66  FMKWAT(//X1H  TwM  Tl'^hS AKI-
          (ill  TM  1O6
      67  HHIlMf  6H, J. f ( J,< ) .T («. .» , I i«l IN. I
      6K  FMKhATf //?'4H  llwt-S MMT Mh
     1C6  CflMf
      70  CUNT
          DM  10?  J=l.j
          D/iY( J )=<).<)
     107  AVt(J)=().l>
          DAY IMF   =11."
                                        143

-------
Table  3.    (Continued)
      OH 1(14  J=1,JA
      IF(N( J)-l ) IDS, HI ,H(>
   «0 K = MJ)-1
      Oil 108  M=1,K
       IF (T( J.M+1 )-T( J ,M) 1? 000,* 001 , JMIO?
 ?()()()  PRINT  6H, J,T ( J ,MAN ) ,T< J,ii ) , I ill IN. I
       GO T'l  108
 2001  H-UNT  66,J,T ( J ,I«AN ) ,T< J,r»i ) , I'l'liM, J YhAK
       (;n TD  lop,
 2002  DAY! J )=IUY( J ) + <(.)< J,M )+'-)( J, 1+1 ) )* )»( 24.0-TI J.K ) )*30.0
       DAYTOT  =f)AYT()T   -K)AY(J)
       AVb(.J ) = ( (I)AY(J) /
       CIIMTFNIIb
       '. = ICUl)t«
       IF (I. )b^O
  •,40  I. =V
       M=ICHtCK
  542  M=9
       K=ICLASS
       K=9
       PRINT  t>53
  553  Ff)RMAT( 1H1,20X,?OHT I'.fc  (HITKi.llw SIIMMAKY/76H  (J'.OT   MONTH    YtflR   Cd
      lOtR    CHtCKtR    DATA KATIKf^    TnrAI. MONTHS  I) I SCHAKGE , 3X ,
      22
-------
Table 3.  (Continued)


      r,o  TO 650
 652  PRINT 555, U,J=L ,M) , IDA YU ),-l=i. ,M) , I AVt(J) ,J=>. ,M) , (N(J) ,J = l,H)
 555  FORMAT 1/6H OATt  , 10110/9H  TOT l-i. (iw . 10HO.O/9H AVt HfcAO,
     110F10.2/10H  l\ll)  POINTS, 10( 17,3X1 )
 650  CONTINUt
      PRINT 660,(l_ ,i_ = l,19)
 660  FORMAT(//6H  DATE , 19(14, 2X))
      DO 670  J=l,JA
      K=N(J>
      1F(K) 666,666,667
 667  PRINT 661, J, (TIJ,'. ) ,!.=!, K»
 661  RlHMATt I^,2X. t 19F6.2 ) )
      PRINT 663, ( S(J,>. ),'. = !, K>
 663 K)RMAT(6X, ( 19F6.2 )//)
      GO TO 670
 666 PRINT 626, J
  626 FORMAT  (I^,/)
 670 CONTINUE
      00 117  J=l,Jft
  117 AVE(J)=0.0
       IF(NPi.OT-99)122,12l,121
  122  M=JA+1
       DO  110 J=1,JA
       DO  110 K=l,40
       T( J.K)=0.0
       S(J,K)=0.0
  1 10  0{ J,K)=0.0
       00  99H J=1,JA
  99H  N( J 1=0
       DO  112 J=M,40
       DO  112 K=l,40
       i, =j-ja
       Td. ,KI=T( J,K)
       Sd. ,K|=S( J,K)
       00. ,K>=0( J,K)
       N(U)=N(J)
   112  CONTINUt
       DO 113 J = M,M)
       DO 113 K=l,40
       T( J,K)=0.0
       S( J.KJsO.O
   113  0(J,K)=0.0
       DO 999 J=M,40
   999  N(J)=0
       GO TO H02
   121  STOP
       tNI)
                                     145

-------
Table 4.  Surface Flow Summary Program.*
Variable Name

   CRF

   IDATE
   IPLOT
   IROGA
   IROGAC

   IROGP
   ISTORM

   IYR
   RF

   RFAC
   ROIA
   ROIP
   ROIAAC

   SOPAC

   SOPML

   SOPPA
   TOTPPA
   TOTPAC

   TOTPML

   XNH4ML

   XNH4AC

   XNH4PA
   XN03AC

   XN03ML

   XN03PA
                  Description

Estimate of the amount of rainfall contri-
  buting to runoff
Date
Plot number
Runoff in gallons per acre
Accumulation of gallons per acre of runoff
  over t ime
Runoff in gallons per plot
Last storm number identification for sampling
  period
Year
Rainfall amount since last sampling period

Accumulation of rainfall over time
Runoff in inches per acre
Runoff in inches per plot
Accumulation of acre inches of runoff over
  time
Accumulation of soluble orthophosphate losses
  in pounds per acre over time
Concentration of soluble orthophosphate in
  mg/1 in runoff
Soluble orthophosphate loss in pounds per acre
Total phosphorus loss in pounds per acre
Accumulation of total phosphorus losses in
  pounds per acre over time
Concentration of total phosphorus in mg/1
  in runoff
Concentration of ammoniacial nitrogen in
  mg/1 in runoff
Accumulation of ammoniacial nitrogen losses
  in pounds per acre over time
Ammoniacial nitrogen loss in pounds per acre
Accumulation of nitrate nitrogen losses in
  pounds per acre over time
Concentration of nitrate nitrogen in mg/1
  in runoff
Nitrogen loss in pounds per acre
      Tile flow summary program is similar except for several
        conversion factors.
                              146

-------
Table A.   (Continued)
              Tn  CAi.cii'.ATb Losses UN  AM ACKt HASIS P> us ACCIHIH A
 C***'_OSStS HY PLOTS     <_ARGt n«AllMA(,H (-XP-KlwhNT- AURORA  M  Y.
 C***PI.IIT  SI7fc=lQOX?Ob httT  =3H,9hOSO  FT = <).9n  ACKfcS
        DIMtNSlDN  TRF(?A) ,mn (?^> ,I«IK;(?^) ,XNM<24 ) ,xMfi(?^) ,snp(24) ,
      CTDTM(24),fMI)h:X(?, l?).C(?4),M(^4),TkHX(l?|,RI!IX( 1? ) , I KOGX ( 1 2 )
      CXNHXI 12) ,XMux(i?),snpxi i?),Tnrpx( i?)
        JCK = 0
        M=n
     t   0(1  bO J=l, ?4
        r/( j)=o.o
        X Ml 13 AC = 0.0
        SHPAC=O.O
        T()TPAC = 0.0
        WflIAAC = 0.0
        THIH;AC = O
        « F AC = o . o
        IF( M.t)
     lb  KIRMATf /,f>X, 
-------
Table 4.   (Continued)
   17 t-OKMATI /.fcX, "TKfcATMENTS ' ,bX. '""ANAGFM(:t\iT = MOOR')         >
   43 CHNTINUt
      GO  Til  (31., lH,31,31,lH,31.lH,l«,31,3l,lH,lH,lP,31,18,31tlH,18,31,
     C31,31,31,lfl,lH),J
   IH >AlRITt(6,l9)
   19 FORMAT I ' + ' ,47X, 'f-tRT F'- ITY= H I GH ' . 1 OX , « N A = IV'OT  AVA P. AH', fc • 1
      Gfl  Tl'l  44
   31 WRJTb(6,32)
   3? RIKMATI ' + ' .47X, ' FbRTI*. 1TY= NORM A'. ' , .1 IIX , ' N A=  NOT AVM'.AHLfc1)
   44 CHNTlNUt                                          •'            I
      WKITttft.20)
   20 FORiiATf//,9X , ' PLOT  ST AT I ST ICS1 .7X, » | ' , l^X , ' ANAi.YT ICA>,   DATA",
     C1HX,•I't!3X,'ACCUMULATIVE   THTA'. S' . 1 ?X, ' I « )
      WKITt(ft,23)
   ?3 FDHMAT!•+•,13011H_))               	
      WR I Tt(ft,24)
   24 FORMAT) 32 X, ' I ' ,49X. • I ' ,4'3X, ' I ' ,/,3?X, < I ' ,49X, •  I ' ,45X, ' | ' )
      WRITt(6,21)                                    '        ;
   ?1 HlKMATI         IX. 'SAMPL t ' ,2X, "TOT. ' , IX , 'CUNT ' ,  1 X, MUNinFF',
     C IX , ' RMNII^F ' , IX, 'MH4-N' , ?X, INU3-M' , IX , « Sf)'_ -M« , IX, 'TUT-P1 ,
     C?X, 'NH4-N' , 1 X. 'MI13-M' , 1 X, 'Stl-.-P1 , IX, ' TMT-P' ,2X, 'TOT' ,2X,
     C iRIINIIhf- ' , IX , 'KDNIIhF • ,?X , iNH4^N' . lx ,,'|\I()3-N' ,?X ,  ' SIX -P ' , IX,
     C 'TI1T-P' ,3X, «C ' )
      w«lTt(6,22)
   2? hnRMAT(2X,'l)ATb',3x,.'PPT.l,lX,'Ppr.',lX,'Ar. .IN.'.lX, 'r,AL/AC' ,1X,
     C 'MG/t. ' , 3X , «Mf,/i. « ,?X, «MG/i. • ,?X , »Mf;/l. ' ,^X , 'I. K/AC  ' , IX,
     C •<_ H/AC • , IX. •'. K,/AC ' , \ X, ". -I/AC • ,?X , ' PPT . • , IX . ' AC  . JNi. • , 1 X,
     C T,Ai_ /AC ' ,2X , 'I. H/AC ' , IX , •'_ K/AC ' ,?X, «'. H/AC ' , IX, 'LH/AC ' )
      WR)Tt(6,23)
   4? CDNTlNllt
   2^ W« 1 Ttl 6,30)        IDATt . I YR ,Rh ,CKF ,H(I|I A, I RMr,A,XMH4MI_ , XNOSMI. ,
     CSDMM'. , TOTPML , XNH4tJA,XNH3lJA, SOPM A , Tl I T PH A ,RFftC,RUI A AC , IROGAC,
     CXNH4AC ,XND3AC .SnPAC ..IIITPAC ,C (J.) ,
   30 R)Ri«'AT(    lX,I4,I2,lX,t-1.?,lX.I-4.?.ix,Hb.3,lX,!6,2X,.F5.3,lX,
     Ct-6.2,lX,F15.3,.lX,t-b.3,?X,t-t).3,]X,(-'5.?,lX,l-'!).3,lX,f->5.3,lX,
     CF5.?,lX,Kb.2,lX,I/,?X,H'5.3,lX,i-'5.?,?X,l;'3.3,lX,h5).3,lX,F3.0)
   41 CMNTINUt
       ICK=fCK+1
      Gfl  TO  S
   50 CIINTINOE
   •>! CONTINUE
   40  STOP
      bN|)
                                       148

-------
Table 5.  Nutrient Balance for Nitrogen (N) and Phosphorus (P) in Pounds/Acre for the
          4/70 - 3/71 Hydrologic Year.
Crop
Fertility
Management
Plot #
Fertilizer N
Crop Residue N
Mineralization N
Total inputs N
Crop
Runoff N
Sediment N
Deep Seepage N
Total Removal N
Loss to Environ.
Denitrification
High
good
2, 11
275
50
50
375
149
3
1
163
316
167
59
Corn
poor
15, 18
275
0
60
335
136
3
0
73
212
76
123
Normal
good poor
10, 21 6
15
50
46
111
134
1
3
23
161
27
0
, 16
15
0
59
74
65
1
1
3
70
5
4
Beans /Wheat
High Normal
good poor good poor
12,
490
22
64
576
98
3
0
223
324
226
252
23 5, 13
490
0
64
554
98
28
3
130
259"
161
393
9, 14
90
18
58
166
98
3
1
41
143
45
23
3, 4
90
0
43
133
86
3
1
15
105
19
28
Wheat
High Normal
good poor good poor
7,
0
15
54
69
73
1
0
19
93
20
0
24 8, 17
0
0
48
48
69
1
6
23
99
30
0
19, 20 1, 22
40 40
19 0
54 34
113 74
68 59
3 1
0 3
8 18
79 22
11 22
34 0

-------
       Table 5 (Continued)
Crop
Fertility
Management
Plot #
Fertilizer P
Crop Residue P
Soil Release P
Total inputs P
Crop P
Runoff P
Sediment P
Deep Seepage P
Total Removal P
Loss to Environ.
Fixation
% of Fixation
Corn
High Normal
good poor good poor
2, 11 15, 18 10, 21 6, 16
65.5 65.5 13.1 13.1
7070
14 14 14 14
86.5 79.5 34.1 27.1
18.8 16.9 18.5 9.6
.116 .437 .047 .140
.149 .273 1.310 .338
.0093 .0083 .0086 .0089
19.07 17.62 19.87 10.09
.274 .718 1.366 .487
67.4 61.9 7.2 24.0
77.9 77.9 26.6 70.4
Beans/Wheat
High Normal
good poor good poor
12, 23 5, 13 9, 16 3, 4
122.3 122.3 34.9 34.9
3030
14 14 14 14
139.3 136.3 51.9 48.9
12.6 12.6 12.5 11.0
.160 .335 .037 .083
.075 .651 .528 .687
.0610 .0079 .0122 .0117
12.90 13.59 13.08 11.78
.296 .994 .577 .78
126.4 122.7 38.8 37.1
90.7 90.0 74.8 75.9
Wheat
High Normal
good poor good poor
7, 26 8, 17 19, 20 1, 22
0 0 8.7 8.7
3030
14 14 14 14
17 14 25.7 22.7
13.6 13.4 12.7 11.3
.150 .334 .185 .078
.071 2.931 .076 1.263
.0335 .0664 .0632 .0306
13.85 16.73 13.02 12.67
.255 3.331 .324 1.372
3.1 -0 12.7 10.0
18.2 0 49.4 44.0
Ui
o

-------
Table 6.  Total Accumulative Runoff and Nutrient Losses from Water

          Quality Research Plots  for a Period from 4/1/70 - 3/31/71.

          Means and L.S.D. Values  for the Main Factors.
                                                              Soluble
                                                              Ortho-
                                                             Phosphate
                                Runoff     NH4-N     NOo-N    (PCh-P)
  Factor                     (Gal./Acre) (Lbs./Acre)(Lbs./Acre)(Lbs./Acre)
Crop


Corn
Beans-Wheat
Wheat
Fertility High

Normal
Management Good

L.S.D.
L.S.D.
L.S.D.
Poor
Crop @ 5%
Fertility @ 5%
Management @ 5%
94,517
109,727
119,625
118,517
97,395
72,596
143,316
1
_ _ _
69,260
0.3755
0.3834
0.5019
0.4528
0.3878
0.4167
0.4239
- - -
- - -
- - -
1.04
7.91
0-72
5.46
0.99
1.06
5.38
4.31
3.52
3.52
0.1851
0.1537
0.1870
0.2555
0.0950
0.1159
0.2346
_ _ _
0.0801
0.0801
   L.S.D. Not  calculated  when non-significant  F  ratio  occurs at  given

   confidence  level.
                                 151

-------
  w
                        Subject Field & Group


                          05B
       SELECTED WATER RESOURCES ABSTRACTS
              INPUT TRANSACTION  FORM
      Dept.  of Agronomy,  College of  Agriculture and Life Sciences
      Cornell University
      Ithaca, N.Y.  14850
     Title
      Management of Nutrients on Agricultural Land for Improved Water Quality
 10
Authors)
 P.  J.  Zwerman
 D.  R.  Bouldin
 T.  E.  Greweling
 S.  D.  Klausner
 D.  J.  Lathwell
 D.  0.  Wilson
                                     16
Project Designation
                                          EPA/OEM Project No.  13020 DPB 08/71
                                      21  Note
 22
     Citation
 23
Descriptors (Starred First)

  *Surface Runoff, *Crop Rotation,  *Nutrient Losses,  Manure and Fertilizer
   Application, Rainfall Simulator
 25
Identifiers (Starred First)
  *Nutrient Losses, *Surface Runoff,  *Crop  Management
 27
Abstract
 A  rainfall simulator was utilized  to  determine the effects of 2, 10, and 20
 year storm frequencies on losses of water,  soil and nutrients from plots subjected
 to different crop rotations, fertilizer schemes and manure applications.  Crop
 rotations, rates of fertilizer and manure were compared.   Simulations were made
 on freshly tilled soil.
 Comparative erosion losses were as follows:  continuous sod   corn - alfalfa
 rotations    continuous corn.  Fertilizer alone tended to  increase runoff, but
 this effect was overcome when fertilizer was used with manure.
 Continuous recording of surface and subsurface flow and subsequent losses of
 nutrients  to the environment was conducted  on larger plots.   Rate and time of
 fertilization determined the plant nutrients lost.   Returning crop residues to
 the soil improved water infiltration,  increasing deep seepage losses.  Proper
 timing  of  fertilizer applications  could control adverse environmental effects.
 Phosphorous inputs into cultural media as it related to algal growth was studied.
 Sustained  concentration determined the biomass of phosphorous.
Abstractor        .
      D. F. Anderson
                               Institution
                                  EPA - Office  of Research and Monitoring
 WR:102 (REV. JULY 1969)
 WRSIC
                        SEND, WITH COPY OF DOCUMENT, TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                                  U.S. DEPARTMENT OF THE INTERIOR
                                                  WASHINGTON. D. C. 20240
                                                            AU.S. GOVERNMENT PRINTING OFFICf: 1972 484-484/150  1-3

-------