' ‘qu‘d‘

 

I LIBRARY. ‘ 1
Michigan State
University

._‘._

fit

 

 

 

This is to certify that the

thesis entitled

WATER TABLE RESPONSE DURING SUBSURFACE IRRIGATION

presented by

KENDALL J. DYKHUIS

has been accepted towards fulfillment
of the requirements for

 

 

/

Date gbhcay‘Jéjfi/f 35/

ajor professor

0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

 

1'... MH- » '. .o ; VA...‘ A
‘ I

. "fin-W. &’ .-
~ 4|

 

IV‘ESI.) BEIURNING MATERIALS:
Place in book drop to
LJBRARJES remove this checkout from
.‘II-zs-IIL. your record. FINES will

be charged if book is
returned after the date

stamped below.

 

 

 

 

 

 

 

 

—_——

mtl 2 0 ‘99"
4,

“3‘9“
:iWéJO'

 

 

 

 

 

 

 

 

 

 

 

WATER TABLE RESPONSE DURING SUBSURFACE IRRIGATION:

by

Kendall J. Dykhuis

A THESIS

Sub-itted To
Michigan State university
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
in

Agricultural Engineering Technology

Department of Agricultural Engineering

1988

ABSTRACT
HATER TABLE RESPONSE DURING SUBSURIACB IRRIGATION
8?
Kendall J. Dykhuis

field experinents were conducted on four sites to
determine water table response during subsurface irrigation.
Hater stage recorders were used to continuously record water
table changes. Manual monitoring of water table depths was
conducted using a blow tube. Results showed linited
horizontal hydraulic conductivity resulted in
evapotranspiration removing saturated soil water faster than
it could move laterally to a monitored position five meters
off the tile lateral on one site. Results of water table
response to subsurface irrigation were presented and comparied
to predictions from the approximate steady state approach for
two sites. Comparison between observed and calulated water
table response showed good correlation on two sites with 10
and 20 a tile spacing, respectively. Saturated horizontal
hydraulic conductivity was considered a major factor governing
performance of subsurface irrigation. All field sites were
not reconmended for subsurface irrigation. Rainfall resulted
in significant rise in the water table. Adequate drainage
must be provided in subsurface irrigation design. Yield
response to irrigation was shown to result in greater yield

increase on granular soil texture than heavier clay soil.

ACKNOWLEDGEMENTS

I wish to thank Dr. Ted Loudon, my major professor, for
his advice, encouragement and thoughtful suggestions. I
thank Dr. George Merva and Dr. Raymond Kunze for the time
and effort they devoted as members of my guidance
committee.

A special appreciation is extended to Advanced Drainage
Systems Inc. of Columbus, Ohio, for the funding of this
project. I want to thank staff menbers of Advanced
Drainage Systems Inc., Mr. Bill Miller, Mr. Steve Helmrich,
Mr. Jerry Richter, and Mr. Dale Treumner, who have
supported the design and concept of subsurface irrigation

TM system.

for farmers using the Irri—Drain

My deepest thanks is extended to the following farmers
who have given of themselves and their resources to make
this project successful: Mr. Brad Singer and Mr. Jake
Ellenbaum of Pigeon, Michigan and Mr. Roland Iott of
Petersburg, Michigan. A special thanks to Mr. John Kemp,
of Sebewaing, Michigan, a tiling contractor freely giving
of his time and equipment. My appreciation is extended to
fellow students who helped with field work, Mr. Phil Brink,
Mr. Buell MaCarther and Mr. Lee Wolfe.

Finally, I want to thank my wife, Virginia, for her

encouragement, love and patience given me over the course

of my studies.

iii

To Joe Boeve

One who has been a friend and
positive influence in my life
by his example in loving and serving
others and wanting their lives to be all

that they can be.

iv

II.

III.

IV.

TABLE OF CONTENTS

Introduction.. ....................................... 1
Objectives.. ......................................... 6
Literature Review .................................... 7
A. Design parameters and procedures ................. 7
B. Plant environment ............................... 21

1. Soil aeration ............................... 21

2. Affect of water table depth on root growth..25

3. Soil nitrogen ........ . ...................... 3O
4. Salts ...... ........... ...................... 32
C. Management of the Water Table.. ................. 33
1. Water table position... ........ . ..... . ...... 34
2. Water table shape.......... ............... ..37
3. Timing of irrigation...... ........ ..........38
D. Energy and Efficiency..... ....... ...............40
E. Sugarbeets under irrigation... ........... .......44
Procedure: Field Investigation ........... . ........ ..48
A. Site selection..... ...................... .......48
B. Site description...................... ......... .49
1. Field Site A............ .................. ..SO
2. Field Site 8.... ..... . .... ................. SS
3. Field Site C...... .......................... 62
4. Field Site D... . . . ..................... 66
C. Water table monitoring... ....................... 70
D. Saturated hydraulic conductivity ................ 73
E. Crop yields and planting inputs ................. 74

V

VI.
VII.
IX.

D Saturated hydraulic conductivity ................ 73
E Crop yields and planting inputs ................. 74
F. Available soil nutrients ........................ 74
G Evapotranspiration .............................. 75
H Precipitation ................................... 76
Results and Discussion .............................. 78
A. Site A .......................................... 78
B. Site B .......................................... 79
C. Manual water table response for Site C

and D ........................................... 96
D. Continuous observations using water stage

recorders on Site C ............................ 112
E. Yield results .................................. 138
Conclusions ........................................ 144
Recommendations .................................... 148
List of References Cited ........................... 149

vi

Table

Table
Table

Table

Table
Table

Table

LIST OF TABLES

Saturated hydraulic conductivity utilizing
velocity head permeameter ...................

Fillable porosity at two depths, Site C & D.

Parameter values for Figure 5.10, where time
(hours) is referenced from initiation of
irrigation and water table elevation (cm)
referenced from center of tile lateral ......

Yield data (kg/ha @ 15.5% moisture,
corn only) ..................................

Plant population at emergence (Plants/ha)...

Crop type, variety, fertilizer and herbicide
program .....................................

Soil ph and available soil nutrients of K2
and P205 ....................................

vii

....92

...103

...108

...139
...139

...140

...140

Figure

Figure

Figure
Figure
Figure
Figure

Figure

Figure

Figure

Figure
Figure

Figure

Figure

Figure

Figure

Figure

h
a:

hbnhufioh
01th

H

.10

.11

LIST OF FIGURES

Schematic of subsurface irrigation-drainage
system .......................................... 4
Schematic of subsurface irrigation-drainage
system ......................................... 15
Schematic diagram of Site A, sandy loam soil...51
Schematic of soil types on Site A .............. 52
Schematic of Site B, clay loam soil ............ 57
Schematic of soil type on Site B ............... 59
Location of harvested plots on Site B (not to
scale) ......................................... 61
Schematic diagram of Site C, sandy clay loam...63
Schematic of water level control structure

and pump setup, Site C ......................... 64
Harvested plots for yield checks on Site C ..... 67
Schematic diagram of Site D, sandy clay loam...69
Illustration of a blow-tube used to monitor

water table depths manually .................... 71
Schematic of Leupold Steven's Water Stage
Recorder for measuring changing water table
elevation ...................................... 72
Water table elevation on five dates, north end,
Site B, early season ........................... 80
Water table elevation on five dates, north end,
Site B, late season ............................ 81
Water table elevation on five dates, south end,
Site B, early season ........................... 82

viii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.10

.11

.12

.13

.14

.15

.16

.17

Water table elevation on five dates, south end,
Site B, late season ............................ 83

Water elevation over the west tile line
monitored at both ends of the field, Site B....88

Water elevation over the east tile line
monitored at both ends of the field, Site B....89

Water table elevations at various times
accross three laterals, furthest from
control structure, Site C ...................... 97

Water table elevations at various times
accross two laterals, furthest from
control structure, Site D ...................... 98

Schematic of water table profile during
subsurface irrigation ......................... 101

Calulated and measured water table rise,
midway between adjacent tile lines for
manual observations ........................... 105

Water table rise from recorded tile lines
verses calculated values, Site C .............. 110

Initial water table movement over and between
tile after onset of irrigation for Site C,
north end ..................................... 114

Initial water table movement over and between
tile after the initiation of irrigation on
Site C, south end ............................. 116

Diurnal fluctuations over and between tile
laterals as effected by ET, north end,
Site C ........................................ 118

Fluctuating water table as effected by ET,
south end of field, Site C .................... 120

Water table between and over tile laterals on
Site C, north end ............................. 121

Influence of rain over tile lateral, Site C,
south end ..................................... 122

ix

Figure 5.18 Decreasing water table due to
evapotranspiration, Site C, north end ......... 124

Figure 5.19 Water table response to ET and pump-startup
over the tile lateral, Site C, south end ...... 126

Figure 5.20 Water table response over tile lateral,
Site C, north end of field .................... 127

Figure 5.21 Lag-time over as compared to between tile
laterals when irrigation resumed, Site C,
south end ..................................... 129

Figure 5.22 Lag-time over as compared to between tile
laterals when irrigation resumes, Site C,
north end ..................................... 130

Figure 5.23 Water table response to rainfall and
drainage over the tile, Site C, south end ..... 131

Figure 5.24 Water table response to rainfall and
drainage, Site C, north end ................... 133

Figure 5.25 Water table drawdown associated with ET,
Site C, north end ............................. 135

I . INTRODUCTION

Many locations in the humid regions of North America
experience excessive soil moisture conditions, especially
during the time of critical preplanting and/or harvesting
Operations. It is also common to have a water deficit in
the soil profile during some part of the growing season.
If the soil water deficit occurs at a critical time in the
crOp's development, it can limit crOp growth and yield.
The objective of tile drainage and subsurface irrigation
through drain tile is to provide a soil environment that
results in optimum conditions for field traffic as well as
for plant growth and crop yield.

The state of Michigan generally has the lowest summer
rainfall of any state east of the Mississippi River.
Drought stress is probable two years out of five for the
summer months June through September. Moisture deficiency
is likely to occur every year during some portion of the
growing season (Lucas and Vitosh, 1978; Linville, 1983).

The state of Michigan has 1.2 million hectares (3
million acres) of soils that would benefit from drainage.
The expected post-drainage yield increases on these soils
ranges from 15 to 30% (Michigan Data, U.S.D.A., 1982).
These increases would primarily be due to the alleviation
of early reason wetness. Because of this and because of

the midseason potential for drought stress, incorporating

subsurface irrigation-drainage system can be an excellent
management tool.

The primary aim of drainage systems in humid regions
like Michigan is twofold: l) to render the soil workable
early in the spring and 2) to remove excess water from the
upper layers of the soil so that air can enter the soil
voids and become available to the roots of plants.

Drainage systems consist of subsurface or surface
components, or both. Subsurface drainage systems are
usually composed of a network of ditches and/or drain tubes
designed to lower the water table by removing excess water
from the soil profile and transferring it to a drainage
outlet. An adequate drainage system in humid regions is
based on a design flow rate. This flow rate is expressed
as the drainage coefficient and is defined as the depth of
water to be removed from the drained area in a unit of time
(cm per day) (Schwab et.al. 1981). Surface drainage should
also be provided by land forming or shaping to remove
excess surface water in a manner consistent with good
erosion control practices. A drainage system, whether
subsurface, surface, or a combination of the two, allows
for trafficable conditions for seedbed preparation in the
spring and harvesting in the fall. It also prohibits
excessive soil water conditions, providing a proper soil
environment for plant growth during the growing season.

Subsurface irrigation may be defined as the process of

regulating the elevation of a shallow ground water table by

artificially adding water underground through the
subsurface drainage system. Water may be fed to the soil
profile by furrows or ditches running parallel to each
other or with clay tile or perforated plastic pipe buried
75 to 120 cm under the surface of the field. A barrier
against excessive losses through deep percolation must
exist in the profile. The barrier may be a relatively
impervious layer which impedes downward movement of water
in the substratum or it may be a permanently high natural
water table which can be manipulated. Theoretically, the
water level is maintained above the impervious barrier at a
selected depth below the soil surface so that the proper
combination of water and air availability to the plant root
is assured (Figure 1.1).

Subsurface irrigation has several major advantages: 1)
an irrigation system and a drainage system are combined,
which allows for substantial savings on initial capital
cost when compared to the cost of two separate systems
(Worm et.al. 1982); 2) there are lower labor requirements
than with sprinkler irrigation; 3) a lower energy cost is
realized than with sprinkler irrigation because no system
pressure is required and soil matric forces as well as the
force of gravity move water throughout the soil profile
with minimal energy inputs; and 4) maintenance
requirements for subsurface irrigation are low (Massey

et.al. 1983).

6E duozzmn

“m9 wuqmmwm ammo mm><4 NZPUEPmum

.xmx \ Exwxxk \kktkkhtx a\\~\\m\\~\\~\\~

 

 

 

J IHA”
303. mm #43
zo....<0.mm_m2m

"' '

~
1/\ x x ”342.45”
mo<mm3m 4.0m

    

 

       
 

 

f: 20— «akin—z.

 

 

 

 

 

m. 3555 206858

:I::::::.:::

km m0 .jduzzm.

Schematic of subsurface irrigation-drainage

system.

Figure 1.1

Dispite these advantages, the practice of subsurface

irrigation has not been rapidly accepted. The primary
reasons seems to be a lack of knowledge about how water
moves during irrigation and drainage phases in specific

soil profiles and knowledge about design for effective

movement of water (Renfro, 1955). With the understanding

of these processes, better subsurface irrigation management

practices may be developed.

II. OBJECTIVES

In order to gain an understanding of the response of
the water table during subsurface irrigation and drainage
and to apply this understanding to subsurface irrigation
and drainage management, the following objectives were

formulated for this study:

1. To measure the water table response to water

input through subsurface tile in different soil

types.

2. To measure the water table response in the

transition from irrigation to drainage modes.

3. To develop management principles which may easily

be applied to on-farm subsurface irrigation.

III. LITERATURE REVIEW

To adequately address the issues of water movement in
the soil with a subsurface irrigation and drainage system,
some knowledge of (a) design parameters and procedures, (b)
plant environment and (c) water table management is
required. This section provides a review of the more

pertinent literature and theory associated with these

topics.
A. Design Parameters and Procedures

Design of a subsurface irrigation system begins with a
site inspection. A site inspection involves the assessment
of the need for drainage, slope of the field, location of
the tile outlet and water input, hydraulic conductivity,
and depth to the impermeable barrier. Luthin (1959)
concluded that it is difficult to transfer subsurface
irrigation-drainage requirements and optimum water table
depths from one location to another. Each location should
be treated as an individual field having its own set of
parameters. Fox et.a1. (1956) indicated that an
appropriate site for subsurface irrigation should have the

following characteristics:

1. The site needs subsurface drainage for the
removal of excess water as it exists in

its present state.

2. The site should be relatively flat, with a surface

slope of less than one percent.

3. There exists on the site a slowly permeable layer
which is relatively shallow, from 1 to 3 meters
below the surface, or the site has a naturally

high water table.

4. The site has an adequate water supply available
which will provide enough water to meet crop

needs during the driest months.

5. There exists a good drainage outlet to let water

out of the system or an outlet where water may be

pumped out.

6. The site has moderate to high lateral hydraulic

conductivity.

Two essential properties to consider for crops being
grown are the range of rooting depths to be accommodated
and different evapotranspiration rates at each growth
stage. These characteristics affect the amount of water
needed and the water table depth to be maintained in
relation to the soil surface. Analysis of climatological
data will show the historical growing season rainfall,
weather patterns, and evaporative conditions which the
design must be prepared to meet.

The most important parameter that influences the

design of a subsurface irrigation-drainage system is a

property called hydraulic conductivity (K). K is the
ability of the conducting medium to transmit a liquid in
response to a head gradient.

Hydraulic conductivity was first defined by the French
engineer Henri Darcy during an investigation of seepage
rates through sand filters (Hillel, 1980). Darcy's law says

that

q = KAH/L ............................. (3.1)
or for one dimensional flow in differential form:

q = -K dH/dx .......................... (3.2)

where q equals the specific discharge rate or volume of
water flowing through a unit cross-sectional area per unit
time (called flux); AH/L equals the hydraulic gradient or
the head drop per unit distance in the direction of flow;
and K is the proportionality factor known as the hydraulic
conductivity with units of length (L) over time (t).

The hydraulic conductivity (K) is affected by physical
properties of the soil profile and by the properties of the
fluid. The soil characteristics which affect K are the
total porosity, the distribution of pore sizes, and the
pore geometry of the soil. The properties of the fluid
which affect K are fluid density and viscosity. The
electrolytic concentration also affects K due to
swelling and dispersion within the soil matrix. Doering

(1965) found from laboratory experiments that the

10

detachment of clay particles and the entrapment of air
bubbles resulted in the clogging of pores, and that a three
fold change in the K value can result from either of these
causes. Criddle and Kalisvaart (1967) pointed out that a
soil profile with layers of silt or clay, even with good
permeability, presents two problems for subsurface
irrigation: 1) such soils generally have a slow capillary
rise, 2) they often lose their permeability under
subsurface irrigation practices. Subsurface irrigation on
such soils may be satisfactory in the beginning, but under
saturated conditions, they often become less permeable or
even impermeable.

Many agricultural soils are not a homogeneous mix, but
can have horizons with varying properties of texture and
thickness, and the hydraulic conductivity may vary with
depth and location.

Once it has been determined from a field evaluation
that a subsurface irrigation system is feasible, the next
step in the design is to choose the spacings of the tile
drainage lines, or the spacing of the lateral ditches. The
objectives of tile and/or ditch spacing are to satisfy both
drainage and subsurface irrigation requirements. The
design must place the drains close enough together to
maintain some minimum water table during peak water use.
The water level in the field must be lowered after a
rainfall or prior to expected rainfall to facilitate rapid

drainage and prevent crop damage due to poor aeration.

11

Therefore, the drain spacing used must:

1. Provide trafficable conditions for timely
tillage and field operations during field

preparation and harvesting;

2. Protect the crop from excessive soil water
conditions by the removal of excessive soil water

during the drainage process;

3. Provide rapid and uniform water input so that
yields are not reduced because of drought stress

(Skaggs et.al. 1973; Skaggs, 1981).

Because tile spacing contributes greatly to the cost
of the subsurface irrigation-drainage system, a knowledge
of the optimum spacing will allow the farmer to
do a cost analysis to determine if the expenses of
construction, when compared with the benefits, justifies
the investment. A site which allows the water to move
laterally through the soil rapidly (sandy loam and loamy
sand) may offer the most significant cost return for two
reasons. First, the tile or open ditch laterals can be
spaced farther apart than laterals in more finely textured
soils. Second, the yield increase due to irrigation is
somewhat greater because the coarser soils do not have the
water holding capacity found in clay and clay loams and

therefore have lower yields without irrigation.

12

At least four methods have been used for determining

tile spacing for subsurface irrigation:

I. The exact approach, which utilizes the numerical
solution of the two-dimensional Richards' equation

for exact boundary conditions;

2. The assumption of the steady state condition

for saturated flow in the soil profile;

3. A (rule of thumb) percent of the spacing required

for drainage alone;

4. Drainmod, a computer simulation program that
considers soil properties, climatological
data, and crop and site properties to predict
the performance of subsurface irrigation
and/or drainage at different tile or open
ditch lateral spacings based on several

years of weather data (Skaggs, 1981).

Exact Approach. Two dimensional water movement in the

 

soil can be characterized by an equation proposed by
Richards (1931) and is an exact approach that uses
appropriate boundary conditions. While this approach has
the advantage of not requiring simplifying assumptions and
of being adaptable to most boundary conditions, numerical
solutions are often difficult to obtain. Tang and Skaggs
(1977) state that it is both difficult and expensive to

characterize effective field values of the soil properties

13
required in the exact approach.

Approximate Steady-State Approach. The approximate
approach is considered less accurate than solutions to the
Richards equation. However, the approximate methods for
predicting water table movement during drainage and
subsurface irrigation is acceptable for most field
situations. The solutions have the advantage of being easy
to apply and require fewer soil property inputs than more
sophisticated approaches (Tang and Skaggs, 1977).

The steady-state condition for determining spacing uses
the approximate method based on the Dupuit-Forcheimer (D-F)
assumptions (Hillel, 1982). These assumptions ignore the
zone in which water moves by unsaturated flow (Bouwer and
van Schilfgaarde, I963; Kirkham, 1958). The assumptions of

D-F are that:

1. Water moves to the drain in a horizontal direction.

2. The velocity at each point in the soil profile is
proportional to the slope of the water table and
independent of the depth while under the influence

of gravity (Hillel, 1982).

Mathematical models using the D-F assumptions assume a
homogeneous soil with a constant hydraulic conductivity and
a constant steady state condition. Hooghoudt (1937), using
the assumptions of D-F, obtained an equation for the

elliptical shape of the water table between the drains

14
for steady state condition (Figure 3.1).
L = ((4Km/q) * (2d+m))0-5 .................... (3.3)

where L equals the distance between the drains, K equals
the saturated hydraulic conductivity, d equals the height
of the drain above the impervious layer, q equals the
quantity of water passing through a unit plane per unit
time, m equals the height of the water table above the
center of the tile lateral, measured at the midpoint
between adjacent tile lines. This equation has been used
frequently in the past (Hillel, 1982).

There have been a number of modifications made to the
steady-state models to account for new understanding of
saturated flow. Bouwer and van Schilfgaarde (1963) modified
Hooghoudt's equation which assumed that the water table
falls without change of shape, and the flux per unit area
of water table is uniform between tile laterals. However,
their assumption that the water table falls without change
of shape is of limited validity. In a ponded condition,
the water table falls faster near the drains than midway
between the drains. Having receded faster near the drains
than midway between the drains, the water table reaches a
position where it falls for some time without appreciable
change in shape (uniform water table movement). As
recession progresses, the water table eventually falls
faster midway between the drains than in the vicinity of

the drains. Thus, the flux in general

15

rues) m>_._.o_E.mmm
.xmxxmxxebxkxk

 
     

.mo. wodnmum ammo

 

’—

 

 

 

 

 

 

 

.B i . .
85 tozemnH m. uoamoem zo_mmumauo

:____::::::_::

...m mo .3323".

 

  

Schmatic of subsurface irrigation

drainage system:

Figure 3.1

l6

varies with distance from the drain. In order to use
steady-state solutions for prediction of the rate of fall
of the water table, which assumes a flux that is
independent of time and distance from the drains, a
correction factor C has been proposed (Bower and van
Schilfgraade, 1963). C accounts for the non-uniformity of
flux per unit area of water table if the water table

changes in shape during recession (or rise).
L = ((4Km/qC) * <2a+m))0-5 ............... (3.4)

C can be determined by calculating the ratio of the
average distance of fall of the entire water table to the
fall midway between the drains for a certain time
increment. Bouwer (1959) has shown that this ratio is
about 0.8 for m/L values from 0.02 to 0.08 and low d values
(Figure 3.1). A C value of approximately 1.0 is indicated
by Childs' (1947) work for higher water tables with m/L
exceeding 0.15. C values of 1.0 are applicable for water
tables that recede uniformly with time. C values greater
than 1.0 can be expected for the initial stages of water
table recession following a ponded case.

Two other D-F assumptions, continuous steady-state
flow into the drain tile without restriction and
homogeneous soil profiles, do not hold true in actual field
situations. Hooghoudt (1937) characterized flow to the
drains by considering radial flow in the region near the

drains and applied the D—F assumptions to the region away

17

from the drains. This analysis has been widely used to
determine an equivalent depth, de, which tends to correct
for convergence near the drains. Moody (1966) when
considering the decrease in the inflow of water near the
tile line as a result of convergence assumed an equivalent
depth, de, at some point above the impervious layer rather
than the actual depth between the impervious layer and the
tile lateral.

Skaggs (1978) used equations developed by Moody (1966)
who examined Hooghoudt's solutions and presented the

following equations from which de can be obtained.

For 0 < d/L < 0.3

d ......... (3.5)

de=
l + d/L*( 8/fl *1n(d/r) - 5:)

in which as = 3.55 - l.6*(d/L) + 21mm.)2

and for d/L > 0.3

de = Ly oooooooooooooooooo (3.6)
8*(ln (L/r) - 1.15)

in which r is the drain radius. Usually «F can be expressed
as <5 = 3.4 with negligible error for design purposes.

The tile's circumferential area is not all porous,
porosity is actually a small fraction of the total area.
The converging water is forced to enter the drainage tile
through a limited inlet area. This results in a

restriction of the water volume entering the drainage tile

18

laterals. To account for this additional convergence loss
as the water approaches the finite number of openings in a
real rather than completely open drain tube, an effective
drain tube radius, re, is defined. Skaggs and Tang (1979)
define an effective drain tube radius, re, such that a
completely open drain tile with radius re will offer the
same resistance to inflow as a real tile line with radius
r. Skaggs and Tang used the work done by Bravo and Schwab
(1975) to define re for a 114 mm (4 inch) corrugated pipe
with slots 1.6 mm wide and 27 mm long as, re = 5.1 mm.

The Dupuit-Forcheimer (D—F) assumptions used in
horizontal saturated lateral steady-state equations assume
a homogeneous profile. Most soils do not have
homogeneous soil profiles but rather have layered profiles
with different properties of texture and porosity. This
results in a non-constant saturated hydraulic conductivity
throughout the soil profile and the Hooghoudt assumptions
not holding true. Compensation for the different soil
profile layers with changing lateral conductivities must be
made (Skaggs, 1979).

The compensation for the lack of homogeneous lateral
hydraulic conductivity through a layered soil profile was
accomplished by measuring the depth and conductive
properties of each layer and taking a weighted average.
The weighted average includes only those layers occupied by
the water table and is called the equivalent hydraulic

conductivity (Ke).

19

Re = Kldl + K202 + x103 ................. (3.7)
11+02+D3

 

where Ke equals the equivalent hydraulic conductivity; K1
equals the lateral conductivity in the first layer and d1
equals the depth of the first layer, K2 equals the lateral
conductivity of the second layer and so on. If the water
table is not positioned in the first layer of soil, then K1
and d1 equal zero.

Bouwer and van Schilfgaarde (1963) found a good
correlation when comparing the approximate method proposed
by Hooghoudt to the more exact numerical solution proposed
by Richards for two-dimensional flow. However, because the
inputs and computations required to use the Richards
equation are difficult, the steady state method using the
D-F assumptions appears preferable for predicting spacing
of drains provided convergence near the drain can be

accounted for (Skaggs, 1978).

Percent of the Spacing. Commonly, the tile

 

spacing used for tile drainage in a given area is the
spacing that seems optimal, based on contractor and farmer
experience with a particular soil type. The spacing chosen
for subsurface irrigation is usually less than the spacing
for drainage. A typical rule of thumb has been to space
laterals for subsurface irrigation at 65% of the spacing

commonly used for drainage.

20

Drainmod. The fourth method which can be used to

 

calculate subsurface irrigation and drainage spacing is a
computer program called Drainmod (Skaggs, 1979). Drainmod
utilizes the soil properties, climatological data, crop
parameters and site parameters to predict the position of
the water table and the total number of trafficable days in
a given period for tillage and harvesting operations. The
model is used to simulate the performance of a chosen tile
or open ditch lateral spacing for subsurface irrigation
and/or drainage over several years of weather data. It
utilizes a quasi—steady state equation to predict drainage
lateral spacings. The Ke concept (equivalent hydraulic
conductivity) is incorporated to account for changes in
hydraulic conductivity for a layered soil profile. The
convergence losses at the drain are treated in the same
manner for subsurface irrigation as for drainage by
equating ho to the sum of de plus the water table
elevation above the center of the drains (yo).

The result from Drainmod is that the spacing for a
subsurface irrigation and drainage system will be predicted
for a determined crop yield performance and trafficable
condition for a particular field, crop characteristics and

geographic location (Skaggs, 1979).

21

B. Plant Environment

1. Soil Aeration

Plant respiration and soil aeration are important
factors that affect crop response when saturated soil
conditions or lack of soil moisture exist (Harris and van
Bavel, 1957a; Grable, 1966). Drainage plays a key role in
the removal of excess water during periods of high soil
moisture content to enhance soil aeration. Subsurface
irrigation plays a role in the addition of water during
conditions of moisture deficit.

The process of respiration by plant roots and
organisms involves the exchange of gases with the soil air.
Harris and van Bavel, (1957b) concluded that root
respiration was the most sensitive aspect of plant activity
in regard to soil aeration. They found that reduction in
respiration activity is the first step in growth-limiting
effects caused by insufficient aeration.

The gaseous exchange of air between the soil and the
atmosphere can occur by two different mechanisms:
convection and diffusion (Hillel, 1982). Convection is the
process of air moving as a mass from a zone of higher
pressure to one of lower pressure. Diffusion is the
movement of each molecular species from a higher
concentration to a lower concentration as a result of
thermal agitation.. An example would be the diffusion of

high C02 concentrations in the soil, produced by the

22

respiration of plant roots and soil organisms, to the
atmosphere where the C02 concentration is less.

The diffusion of gas can take two different forms. It
can occur within the soil profile partly in the gaseous
phase and partly in the liquid phase. Diffusion through
the air—filled pores maintains the exchange of gases
between the atmosphere and the soil, whereas diffusion
through water films of various thicknesses maintains the
supply of oxygen to, and the disposal of C02 from, live
tissue such as root hairs. The rate of diffusion, however,
is greater in the gaseous phase than in the dissolved water
phase.

There has been some discussion about the process and
exact amount of 02 diffusion which takes place in the root
zone. Hillel (1980) describes the diffusion process by
Fick's law, where diffusion is a flux (mass diffusing
across a unit area per unit time). Lemon and Erickson
(1952) proposed a means of measuring the potential 02
diffusion rate supplied to the root zone with a platinum
electrode. Oxygen must diffuse through a film of water
surrounding the platinum electrode wire much as oxygen
diffuses to a root; thus the measurement approximates the
rate at which oxygen can continually be supplied to a root.

The amount of 02 diffusing to the root zone tends to
vary from one researcher's findings to another. Erickson
(1965) concluded that there is a certain threshhold

oxygen diffusion rate below which plants cannot survive.

23

There is then a response curve where increasing the oxygen
diffusion rate results in an increase in plant growth. The
critical 02 diffusion rate for most plants was
approximately 35-40x10'8 gms/cmZ-min, and partial
suffocation may result below this critical level. Erickson
did not correlate the diffusion rate with any stage of
growth in the plant. Williamson (1964) set up an
experiment using lysimeters in field installations to
measure the rate of 02 diffusion in the soil. Water tables
were maintained throughout the growing season at intervals
ranging from 150 to 762 mm below the soil surface. The
soil type was a fine sandy loam. His conclusions were that
the highest yields were obtained at 02 diffusion rates of

2-min for sweet corn and dwarf

approximately 15x10”8 gms/cm
field corn when soil moisture was not limiting in a growing
crop.

The review given by Grable (1966) made clear that the
effect of temporary 02 deficiency caused by flooding
depends not only on plant species but also on the
physiological stage of growth, time and duration of water
logging, light intensity, fertility of the soil, and
temperature. A growing plant needs a critical amount
of 02 to maintain metabolic function. Williamson and Kris
(1970) and Unger and Danielson (1965) found that a reduced
0; supply rather than a buildup of C02 may be responsible

for reduced growth of young corn plants under poorly

aerated soil conditions.

24

The location of a static water table affects the rate
of Oz diffusion into the root zone. Hiler et.a1. (1971)

found that with a static water table at 300 mm, the 02

2-min in the

diffusion rates never exceeded 20x10‘8 gms/cm
root zone during the growing season. Williamson (1964)
concluded that 02 diffusion rates with a 150 mm water table
depth were below 5x10.8 gms/cmz-min. Under these
conditions, sweet corn and dwarf corn yields were reduced
65-75 percent respectively. Williamson also found that
there was a tremendous increase in the 02 diffusion rates
above 15x10"8 gms/cmZ-min where the soil moisture was not
limiting growth.

It has been noted that light intensity and its
associated photosynthetic and transpiration demand can have
adverse effects on plants when a high moisture-low aeration
condition prevail (Erickson, 1965). Lemon and Wiegand
(1962) showed that there is also a characteristic increase
in the critical 02 concentration as the temperature
increases. Thus, as light intensity and higher air
temperatures put more demand on the metabolic activities of
the plant, plant respiration increases. The higher
respiration increases the utilization of 02, increases the
C02 concentration, and increases the 02 absorption demand
at the root surface. Since biological processes are more
temperature sensitive than physical processes such as
diffusion, diffusion is more likely to be the limiting

process as the temperature is increased in the absorption

25

of O2 by the plant. This becomes understandable when one
realizes that the diffusion path from soil surface to root
surface is completed via the water films surrounding the
root hairs. This film may constitute only a fraction of
the total length of the path, but its effective length is
increased more than 10,000 fold by virtue of the smaller
diffusivity of 02 in water than in air. Letey and Stolzy
(1967) concluded that a film thickness of 0.08 to 0.05 mm
would limit root growth in media containing 20 to 50
percent air porosity and atmospheric Oz. Hillel (1982)
discusses the different methods of measuring aspects of
soil aeration.

It is thus evident that the degree of soil
aeration affects the growth potential of growing crops.
The water logging of the soil profile increases the
effective thickness of water films surrounding the plant
roots. The water film thickness affects the effective
length of diffusion through which 02 must travel by as much
as 10,000 times when compared with diffusion through air.
Temperature and the high moisture-low aeration condition

have an adverse effect on crop growth and yield.

2. Affect of Water Table Depth on Root Growth

How a plant grows and the environment in which the
growth occurs affect the outcome of the plant and its
yield. The distribution and depth of roots depend on many

factors such as plant species, stage of growth, fertilizer

26

distribution, tillage treatments, physical barriers such as
hardpans, chemical barriers such as acid subsoils, and the

soil water regime during root development (Allmaras et.al.

1973).

Roots grow in the upper crust of the soil profile and
penetrate and spread out as they mature. The depth and
concentration of the roots is important information,
useful in establishing the proper placement of the static
water table in subsurface irrigation. Bloodworth et.al.
(1958), in their work on root distribution of crops using
undisturbed soil cores, found that 90 percent of grain
sorghum roots were concentrated in the top 300 mm of the
soil. Mengel and Barber (1974) found the highest
percentage of corn root distribution, 80 percent, to be in
the upper 400 mm of the soil profile. They also observed
little evidence of root growth limitation by moisture or
aeration stresses on corn grown in a silt loam soil having
tile one meter deep spaced twenty meters apart. Williamson
and Kris (1970), however, concluded that the root zone will
be restricted by the water table. A study done by
Williamson et.al. (1969) with millet indicated that total
root weight increased with increasing depth of the water
table to a maximum at 610 mm. Although most roots were in
the top 150 mm, the highest shoot yields were obtained from
tanks with water tables at 760 to 1020 mm.

Chandra (1973) concluded that final yields were not at

all related to the amount of root growth near maturity but

27

were strongly related to the total root length as well as
to the depth distribution of roots during the

vegetative phase of growth. Evans et.al. (1985) concluded
that crop yield reductions of more than 50 percent may
develop from stress caused by excessive soil-water
conditions on poorly drained soils. Greater depth to the
water table apparently permits more soil aeration and
nutrient uptake for greater root development. Passioura
(1981) found that the root system of cotton growing in the
subsoil of a drying profile grew much faster than those
growing in a well-watered profile. The total length of the
drought root system was substantially greater than the
well-watered one. The most likely explanation for this was
that drought affects the growth of the shoot more than it
does photosynthesis, so that the amount of assimilation
available for root growth was thereby increased. The
increased root development helps match the plant's water
supply to the evaporative demands on its leaves.

High water table affects the availability of moisture.
Kramer (1965) showed that having a water table too high or
in a flooded condition can hinder upward water movement in
the plant root system. He found that saturating the soil
with water caused a reduction in water movement through the
root system of tobacco to 60 percent of the original rate
in an hour and to 25 percent in three hours. Injury due to
deficient aeration when flooding occurs for longer than 24

hours can result in serious and often irreversible symptoms

28

such as epinasty (the downward curvature of the petioles),
yellowing, and finally death of the lower leaves. Kramer
(1965) and Williamson (1964) found that corn plants grown
with a 152 mm (6 inch) water table were slightly chlorotic
and somewhat smaller than were plants grown with deeper
water tables. It was also noticed that the corn plants
remained chlorotic throughout the growing season, and that
the smallest diameter corn stalks were associated with
plants grown at a 152 mm (6 inch) water table depth.

The optimum water table depth is dependent on the
crop, stage of growth, and soil type. Williamson and Kriz
(1970) related yields for a variety of crops to the
position of varying piezometric water table depths in
varying soil textures. They compared water table depth to
the relative yield, where 100 percent relative yield
equaled maximum potential yield if no crop input parameters
were limiting. They concluded that the severity of injury
for corn as it was actively growing was dependent on the
growth stage and the time of year during which excessive
high soil moisture took place. Purvis and Williamson
(1972) concluded that the flooding of corn for one day had
no visible effect on growth and yield. However, the
flooding of corn for more than one day in the early growth
stages caused reduction in yields.

The desire to look quantitatively at stress due to
excess water or lack of water has resulted in the

development of a number of models. The concept of Excess

29

Soil Water (SEW30) was discussed by Wesseling (1974) and
Bouwer (1966). It was originally defined by Sieben (1964)
to evaluate the influence of high fluctuating water tables
during the winter on cereal crops. The value of SEW30 is
equal to the length of time in days the water table stayed
in the top 30 cm of soil. Computing a range of SEW3O
values and matching yield reduction to SEW30 values, the
yield loss could be predicted for the crOp grown. The use
of the SEW30 concept assumes, however, that the effect on
crop production of a 50 mm water table depth for one day
duration is the same as that of a 250 mm depth for five
days. This seems unlikely as pointed out by Wesseling
(1974).

At the other extreme is a water table too deep to
allow the majority of the roots adequate moisture. The
plant can compensate somewhat for this condition. Reicosky
et.al. (1972) indicated water uptake over a static water
table may not be related to root distribution. They found
that a small amount of roots near the capillary fringe
absorbed most of the water. This may be compared perhaps
to trickle irrigation where water is placed at a very
specific location and reaches only part of the root system.
However, forcing a plant to extend its rooting depth and
distribution can have positive effects. More nutrients
become available to the plant and less irrigation water is
needed because the volumetric soil water content available

to the plant is increased. Hiler (1969) used early season

3O

drought stresses to stimulate downward development of the
root system.

Hardjoamidjojo and Skaggs (1982) incorporated the
Stress Day Index (SDI) proposed by Hiler (1969) as a water
management parameter in the simulation model, DRAINMOD, to
quantify the effects of too little or too much water on
corn yields. They compared actual yields with predicted
yields. The predicted and measured results were in good
agreement. Based on the predicted yields, DRAINMOD could
then be used to design a subsurface irrigation and drainage
system to decrease water stresses that hinder production.

It is apparent that a water table needs to be
positioned to allow for maximum root length and depthwise
distribution to increase nutrient uptake and soil aeration.
The water table should also be positioned in the soil
profile so that upward flux of the water can supply
sufficient water to the root zone. Flooding the root zone
can result in a decrease of water movement into the root
system and can produce a chlorotic effect in the corn

plants and thus a reduction in yields (Williamson, 1964).

3. Soil Nitrogen

The flooding and/or water logging of soils leads to a
series of undesirable events involving soil nitrogen.
Fertilizer effectiveness is reduced, nitrification stops
and denitrification takes place rapidly. Allison (1965)

concluded that of the number of pathways nitrogen can be

31

lost in the soil profile, most are minor except
denitrification and leaching.

Denitrification is the process of nitrogen, in the form
of nitrate (N03), being lost from the soil as nitrogen gas
(N2). This loss of nitrogen due to denitrification is
attributed to the activities of anaerobic denitrifying
bacteria that utilize oxidized forms of nitrogen. The
bacteria demand oxygen to carry on their metabolic
processes. When the 02 supply is decreased because of a
high water table condition, the denitrifying bacteria
remove 02 molecules from the available nitrogen (N03) ions.
Nitrate, upon losing the 02 molecule, becomes free nitrogen
and escapes into the atmosphere. Williamson and Willey
(1964) found that the beneficial presence of N03 lasts for
only a few weeks during a normal growing season. In their
study, denitrification and leaching of nitrate resulted in
deficient nitrate within four weeks after nitrogen
application. Alternate submergence and aeration has the
effect of rapidly depleting the soil of available nitrogen
through the denitrification process (Patrick and Wyatt,
1964).

Leaching loss of nitrogen from the soil system is
dependent on the nitrate ions available and the quantity of
water passing through the soil profile. Leaching losses of
70 percent of the applied nitrogen have been reported under
conditions of overapplication of sprinkler irrigation

(Johnston et.al. 1965).

32

In a study on water table control on organic soils,
yields of onions, carrots, potatoes, peppermint and corn
were reduced with a water table less than 400 mm, and crop
response to nitrogen was also limited with the water table
less than 400 mm (Harris et.al. 1962). Because the process
of denitrification with a 400 mm or higher water table
resulted in a limited crop response to nitrogen, it stands
to reason that controlling the depth of the water table at
a depth deeper than 400 mm is important in maintaining the
nitrogen level in the root zone of the above plants in

organic soil.

4. Salts

Consideration should be given to the concentration of
soluble salts and sodium in the water supply and soil
profile where they pose a problem during irrigation. The
practice of subsurface irrigation can promote the
concentration of salts in the surface soil (Renfro, 1955).
The salts accumulate at those depths where roots absorb
water for the evaporatranspiration process. The osmotic
pressure of the soil solution within the capillary zone
increases as salts accumulate. This reduces the
availability of water in the root zone area.

When adequate drainage is maintained, late season
winter and spring rains, snow melt or high annual rainfall
can wash much of the surface accumulation of salts downward

in the soil profile (Kriz and Skaggs, 1973).

33

Concentrations of salts are measured by the conductivity of
the saturated extract (Williamson and Carrekar, 1970;
U.S.D.A. Handbook No. 60, 1954). When the conductivity of
a saturated extract is more than four mmhos per cm in the
soil profile, plant growth (or water uptake) may be limited
for some crops (Long, 1965).

Therefore, soluble salts and sodium can be a problem
in areas low in annual rainfall and snow melt. However,
because of the amount of annual rainfall and snow melt in
the state of Michigan, salt accumulation during subsurface

irrigation has never been identified as a problem.

C. Management of The Water Table

A subsurface irrigation and drainage system can
eliminate part of the risk associated with growing a crop.
Risk associated with excess soil water and drought
conditions can be greatly decreased because of the ability
to irrigate and drain excess soil water. Good management
of such a system will not only increase average yield but
also reduce the yearly variations in yields. The position
of the water table, the timing of water table establishment
at the base of the root zone, and the shape of the water

table are critical factors determining the success of water

table management.

34

1. Water Table Position

The proper position of the water table is a function
of the management ability of the farmer, soil texture, and
type of crop being grown.

Williamson and Kriz (1970) concluded that
coarse-textured soils require a higher water table for
optimum yields than do fine-textured soils because moisture
will move higher above the water table in a fine-textured
soil by capillary action. This upward flux of water at any
point in the soil is dependent on the depth of the water
table, the evapotranspiration rate of the crop planted
(i.e., plant water uptake rate), soil matric potential, and
soil texture. The deeper the water table, the farther
the water must move through the capillary zone to reach the
roots. Skaggs (1981) concluded that the water table
elevation which should be maintained during subsurface
irrigation depends on the depth and distribution of plant
roots and the rate water can be transmitted upward from the
water table into the root zone. A deep water table may
place the water out of reach of the crop roots. A shallow
water table will produce poor soil aeration, decreased
plant growth, and potentially decreased yields.

Recommendations for optimal water table position vary
among researchers. Williamson and van Schilfgaarde (1965),
Williamson and Kriz (1970), and Vissir (1958) reported that
the static water table depths giving Optimum yields are

generally 760-860 mm for corn in fine sandy loams and loam

35

soils. Jensen et.al. (1980) reported that for most soil
types the water table should not rise closer to the soil
surface than 500 mm for most crops and should fall to the
design water table depth within one or two days if the
water table rise was due to rain or overpumping.
Williamson (1964) noted a decline in soybean yields when
the water table was at a depth of 760 mm from the surface
as compared to 460 mm. It was further noted that this was
due to a drying out of the top layer of soil where a
majority of the soybean roots were located. Williamson and
van Schilfgaarde (1965) found in four experiments that
soybeans grown in Norfolk and Seneca fine sandy loams and
Bayboro loam soils were found to have a maximum yield at a
water table depth of 460 to 620 mm. Woolley (1965)
mentions a conclusion of Williamson that most agricultural
crops grow well in the upper 375 mm of soil if it is well
drained.

Two approaches have been used to manage a subsurface
irrigation system. The first method is to maintain a
constant water table at some depth below the soil surface.
The water table is allowed to fluctuate only one or two
inches and is easily controlled by the use of a float
switch which turns the irrigation pumps on and off.
Management is greatly simplified. One disadvantage of this
method is that limited storage is available for rainfall if
the water table is held too high. As a result, most

rainfall that occurs during the irrigation season drains

36

from the profile. Williamson and Kris (1970) looked at
this approach and related yield to a variety of constant
water table depths. They considered approximately thirty
different crops and computed percentage of the maximum
potential crop yield achieved with various water depths on
a variety of soil types.

The second method takes the approach of raising the
water table to the desired level and then shutting off the
pumps. The water table is allowed to decrease through
losses of seepage and evapotranspiration to some allowable
limit (i.e., 203.2 to 457.2 mm of fluctuation). Once the
limit is reached, the irrigation pumps are turned back on.
This approach has greater potential to store and utilize
rainfall that may occur when the water table is in the
deeper range. Less water may be needed as a result, and
pumping costs are minimized. Skaggs (1981) reported that
when using this method the limit to which the depth of the
water table is allowed to fall must be managed closely.
Problems will occur if the water table is allowed to fall
too far which may cause the soil profile to dry out. The
time necessary to raise the water table once again can
create a lag time between 80-100 hours for 18 meter tile
spacing on a sandy loam soil. (Lag time is the difference
in time between the time the pumps are started to the time
it takes to get a response or rise in the water table
midway between the tile laterals.) A large

evapotranspiration rate could cause the lag time to be even

37

longer, possibly stressing the crop. If the tile lateral
or ditch lateral spacing were further apart than
recommended by the design, the lag time would be still
longer because the distance the water must travel is
greater.

Evans et.al. (1985) recommends putting the water table
at the greatest depth for either method used. This level
will provide the greatest storage potential and the most
efficient use of rainfall. It is important not to let the
water table drop too far because of the lag-time it can
take to pump the water table to an effective level for the
crop to use it. The water table depth should be

monitored constantly.

2. Water Table Shape

An understanding of the shape of the water table
during the drainage and subsurface irrigation modes can
help the irrigator monitor and manage his irrigation
system. This can be accomplished by placing monitoring
stations evenly spaced between two drainage tile lateral
and/or lateral ditches (Skaggs et.e1. 1972). PVC tubing
is often used and placed at a depth of 1.0 to 1.2 meters.
The changing water table can then be measured in reference
to its depth from the surface.

It is a proven fact (Skaggs 1981) that the Dupuit-
Forchheimer assumptions hold true and that the water table

assumes an elliptical shape for the evapotranspiration

38

condition. When the subsurface irrigation-drainage system
is in the drainage mode, a water table is created that is
lower at the tile/open ditch lateral than at the midpoint
between laterals (Figure 3.1). Neal (1934) and Tisdall
(1942) give specific data from a silt loam soil showing a
curvilinear water table between drains spaced 13.7 to 38 m
apart with the average water table 183 mm lower at the
drain than at the midpoint.

If the irrigator/manager shuts the drainage system off
when the water table midway between the drains reaches a
certain depth, the water table there would continue to
recede. The lower water table over the laterals would be
filled by water midway between the tile laterals due to the
hydraulic gradient toward the tile or ditch laterals. If
subsurface irrigation were resumed, the water table would
not rise immediately midway between laterals. Skaggs
et.a1. (1972) showed from field studies that for initially
draining profiles, the water table elevation midway between

drains continued to recede after water input was initiated.

3. Timing of irrigation

An important consideration is when to establish and
when to release the water table for subsurface irrigation.
Because crops can vary in their water consumption rate
throughout the growing season, the irrigator must know at
which times water demand will be greatest. Williamson

and van Schilfgaade (1965) showed that as soybean plants

39

matured, the soil moisture decreased over time at
established water table depths of 305, 457, 610, and 762
mm. Lucas and Vitosh (1978) report that the greatest
benefit from irrigation for corn is prior to tasseling and
then through the pollination and silking stages.

It is important that the irrigator not let the static
water table fall more than a few feet below the ground
surface or let the soil profile dry out too much before
starting to establish the water table at the desired level.
A deep water table will require a longer time to pump the
water table up to the root zone, and a dry soil profile
will require a large volume of water per unit of rise of
the water table height. Evans et.a1. (1985) report that
a drying soil profile decreases the hydraulic conductivity
drastically. They report that a soil which normally
required two to three days to move water from the drain to
midway between the drains in a wet soil, may take two to
three weeks to travel that same distance once the soil drys
out. Skaggs et.al. (1972), observed that from the time the
draining of the soil profile was stopped and the pumping of
water resumed, the time-lag was as much as 2.5 days before
the water table began to rise at the midpoint between
lateral tile lines.

Information is lacking on the end of season
discontinuation for subsurface irrigation. Sprinkler
irrigation can be discontinued on corn when the grain is in

full dent stage and moisture is 35 percent. At this stage

40

the black layer forms at the base of the kernels.
Subsurface irrigation may be discontinued much earlier.
This is because the water table has provided and will
continue to provide moisture to the soil profile. This
soil water will provide enough moisture to bring the
growing plant through to maturity, depending on soil type,
for several days to a few weeks. A heavier soil will
eabsorb and hold more plant-available water than a lighter
:3éandy soil. This is a decision the irrigator/manager must
nnéake based on his knowledge of the soil type, maturity of

t;r1e crop, and expected weather conditions.

D. Energy and Efficiency

The energy cost and the efficiency of an irrigation
system are important considerations when choosing which
type of system to use. Subsurface irrigation offers
Ipc>tential for energy savings when contrasted to sprinkler

irrrigation, because it does not require delivery of water
Lnnder high pressure but employs the use of gravity to move
the water. The type of water supply, ground water or
surface water, affects the total hydraulic head that a
system must generate. Energy required for pumping water
makes up a large portion of the total agricultural energy
requirements where irrigation is used. Irrigation energy
requirements may be reduced by a) improving pumping plant
efficiencies, b) improving irrigation efficiency so that

less water is required, and c) lowering the pressure of the

41

system (Smith, 1983).

Gilley and Watts (1977) illustrated that seasonal
pumping energy required for irrigation is dependent on the
volume of water pumped and the hydraulic head against which
it is pumped. The energy required for pumping irrigation

water may be calculated from the following equation:
95 = CDH ............................. (3.8)

where PE = pumping energy required in kw—h/ha (kilowatt
hours per hectar); C = 0.2717 (conversion factor); D =
depth of water pumped (cm); and H = total hydraulic head
against which water is pumped (m).

Larson and Fangmeier (1978) found that energy
requirements for sprinkler irrigation were five to twelve
times greater than for surface irrigation when the water
supply was a surface source. When the source of irrigation
water was ground water, sprinkler irrigation energy
requirements were six to seven times greater than for
surface irrigation. Smith (1983) found that subsurface
irrigation required only about 6 to 9 percent of the energy
required for sprinkler irrigation systems operating at 340
and 690 kpa (50 and 100 psi) respectively. When a deep
well water supply was assumed, subirrigation offered a 20
to 40 percent energy savings for two of three sites and had
equal energy requirement with sprinkler irrigation for a
third site.

The amount of water used during the growing season with

42

subsurface irrigation varies with crOp type, stage of crop
growth, depth to the impermeable layer, seepage losses, and
location. Knowledge of the consumptive use of water on a
daily or weekly basis is necessary to manage the water
application rate to the crop.

Massey et.al. (1983) and Strickland et.al. (1981) each
studied the energy and water use of subsurface and center
pivot irrigation. It was noted that the subsurface
irrigation used 40 to 80 mm more water than the center
pivot system. A number of reasons account for why a
subsurface irrigation system would use more water. Natural
rainfall had reduced storage availability and increased
drainage and surface runoff over that of a sprinkler
system. This made less efficient use of natural rainfall
and required more water to be pumped. Seepage losses occur
vertically and laterally. A significant volume of water
must be pumped to raise the water table near the root zone.

Water table location in the soil profile also affect
cost and water use. Benz et.al. (1981) found small
increases in water table depth beyond the optimum depth,
1.3 m, increased the water and thus energy requirements on
corn and sugar beets. By allowing the water table to drop
0.2 to 0.3 m below the optimum water table depth, 260 mm of
water must be pumped to raise the water table back to the
optimum depth for maximum corn dry matter and grain yield.
Subsurface irrigation requirements can be reduced by

controlling the system such that the midpoint water table

43

is allowed to fluctuate within limits. Loss of water by
evapotranspiration causes the water table between the
drains to be drawn down for a period of time each day
before beginning to rise later (usually during the night)
as the inflow rate exceeds evapotranspiration and seepage
losses. A model based on a numerical solution to the
Boussinesq equation showed that the irrigation requirement
was decreased by about 7 percent when the water table was
allowed to fluctuate compared with a constant water level
control for five simulations (Smith, 1983). Water input
rate should be sufficient so that the water table will
rebound to its initial level before the evapotranspiration
demand for the next day begins.

Subsurface subirrigation can therefore offer
substantial savings in energy cost when compared with
center pivot irrigation systems. This is due in part to
the fact that gravity and/or low pressure pipe flow are
utilized to move water into the soil profile rather than
high pressure. This may be offset, however, by the amount
of water needed to be pumped. A greater amount of water
may be needed for subsurface irrigation versus center pivot
irrigation. The depth of the impermeable barrier, soil
surface, seepage losses, and possible runoff from rain
must be evaluated. Having an impervious barrier which is
very deep and allows the water table to fall substantially
below the root zone and subsurface tile can result in

additional water being needed to bring the piezeometric

44

water surface within the root zone and therefore increases

the cost of subsurface irrigation.

E. Sugarbeets Under Irrigation

Sugarbeets (Beta Vulgaris) have been a major crop in
many areas of the United States for production of sugar.
The objective of subirrigation of sugarbeets is to increase
the weight and sugar content of the beets.

Factors which affect beet sugar content and growth are
plant population, nitrogen levels in the soil profile, and
available moisture. Robins et.al. (1956) and Loomis and
Haddock (1963) found that sugar content was significantly
reduced at increasing nitrogen fertilization rates.
Haddock (1947) found that additional nitrogen increased the
weight yield of the sugarbeet root and simultaneously
decreased the sugar percentage. Deficiencies in either
nitrogen or moisture resulted in a lower rate of root
growth and an increase in sucrose concentration. The
increase in sucrose concentration when both nitrogen and
water are deficient is due to a reduced rate of dry matter
accumulation. Loomis and Haddock (1967) and Henderson et.
a1. (1968) indicated that nitrogen deficiency and drought
stress reduced growth and increased (on a fresh basis) the
sucrose concentration in roots. However, there is a need
to manage soil nitrogen levels to achieve significant root
weight at harvest.

Archibald and Haddock (1959) found that plants kept

45

moist all season by sprinkler irrigation resulted in the
highest yield. He also found that beets kept moist all
season tended to mature early and were generally higher in
sucrose than those from plots which were dry for an
extended period of time. Loomis and Haddock (1967)
reported that sucrose yields are not increased and may be
reduced significantly by single, brief cycles of wilting
and almost invariably are lowered by repeated wilting
apparently due to reduction in leaf area and rate of
photosynthesis. Extended periods of drought preceding
harvest resulted in substantial reductions in sucrose
percentages, and it appears advisable to continue with
moderate irrigation until shortly before harvest. Beet
purity and sucrose yields were increased by nitrogen
deficiency but not by moisture stress. Archibald and
Haddock (1959) showed that the lower the available nitrogen
supply of a soil, the more seriously will excess irrigation
water depress yields. The more abundant the available
nitrogen supply in a soil the smaller the reduction in
yield will be from excess water application.

A recommendation about when to irrigate sugarbeets is
made by Loomis and Haddock (1967), Haddock (1947) and
Brewbaker (1934). They recommend irrigation practices be
directed toward rapid stand establishment and early
attainment of a full leaf canopy. Brewbaker (1934) found
that larger beet yields were obtained with an early first

application and a very late last application of water, with

46

applications made at biweekly intervals using overhead
irrigation during the summer. Midsummer irrigation was
sufficient to avoid periodic water stress which might
reduce root growth and the photosynthetic effectiveness of
the foliage canopy.

Sugarbeets are a deep rooting plant with a large tap
root. Water is extracted most rapidly from the first 0.66
m of soil, but with deep, well drained soils, appreciable
amounts of water will also be obtained from depths of 1.22
to 1.52 m. Loomis and Haddock (1967) state that 60
percent of the total seasonal water used is commonly
acquired from the top 0.66 m of soil. Henderson et.al.
(1968) found that with a water table at approximately one
meter, the top 0.66 meters of soil became very dry with
subsurface irrigation. Most of the water used by the crop
comes from stored moisture in the soil. When this is
largely depleted, nearly all of the water is supplied by
upward rise from the water table. Reichman et.al. (1977)
found in his test with a sandy loam soil, that a shallow
water table at approximately 0.86 to 1.10 meters provided
all the moisture the sugar beets needed (illustrated by no
increase in root weight when the subsurface irrigated plot
was compared with an overhead irrigated plot). Follett et.
a1. (1974) found that the effective root zone for beets in

a sand soil in irrigable areas of North Dakota is only 0.61

to 0.92 meters.

47

In conclusion, sugarbeets need an adequate supply of
soil moisture and soil nitrogen to maintain good yields of
sugar and weight of beets harvested. Water management
should be aimed at rapid stand establishment and early
attainment of leaf canopy. Soil moisture should be readily
available until shortly before harvesting. Establishing a
water table using subsurface irrigation at approximately
one meter to 0.61 meters will provide the moisture the

growing crop will need in a sandy loam to loamy sand soil.

IV. PROCEDURE: FIELD INVESTIGATION

A. Site Selection

This study was prompted by the interest of a major
drainage tubing manufacturer, Advance Drainage Systems,
Inc. (ADS)*. They were interested in field measurements of
the rate and uniformity of water table elevation response
during subsurface irrigation. Names and locations of
selected farmers who were operating systems were provided
by the company. Four systems were selected for study on
the following farms:

1. The C. Iott farm, near Deerfield, MI;
2. The J. Ellenbaum farm, near Pigeon, MI;

3. The B. Singer farm, near Unionville, MI
(two systems).

The purpose of the study was to determine the water
table response while irrigating through subsurface tile
drains and to study water movement under subsurface
irrigation and drainage conditions. Schematic sketches of
the field installations are shown in Figures 4.1, 4.3, 4.6
and 4.9. Criteria for selecting the fields to be studied
were:

Advanced Drainage Systems, Inc.
3300 Riverside Drive
Columbus, OH 43221

48

49

1. To investigate system performance in a range of

soil types;

2. To limit systems selected to systems at least two
years old to minimize soil disturbance effects
resulting from the installation of the subsurface

irrigation and drainage system;

3. To select farmers who would keep records regarding
system operation including such things as when
irrigation pumps were turned on, how long the
pumps were run, when the system was put into the

drainage mode, etc.

B. Site Descriptions

The predominate soil series in the study regions are
classified in Michigan Soil Associations 63,69,71 and 74
(Soil Association Map Of Michigan, 1981). The soils of
these associations were developed under poor natural
drainage conditions from loam, clay loam, or sandy loam
parent material. Strong gleying* and/or mottling
is associated with these soils and naturally poor drainage

is a principle hazard to crop production.

 

* Gleying results from the chemical reduction of iron in
an anaerobic soil condition and is visually recognized by
the gray color in the soil profile. It signifies a high

water table condition.

50

These can become some of the most productive in the state
of Michigan when subsurface drainage is utilized to control

excessive moisture and improve soil aeration.

1. Field Site A

Site A (C. Iott farm) was in a field two kilometers
southeast of Deerfield, Michigan (Figure 4.1). Total
tillable area consisted of approximately 12.7 ha (31.5 ac).
One half of the area was subsurface irrigated, and the
other half was a tile drained but non-irrigated area.

The legal description of the irrigated plot was the N 1/2
of the SE 1/4 of the SW 1/4, and the non-irrigated plot
was the S 1/2 of the NE 1/4 of the SW 1/4 of section 18, R.
6E. T.7S, Monroe County (USDA, Soil Survey, Monroe County,
1980).

The field drained into a ditch bordering the east
boundary. The ditch in turn drained into the Raisin River.
The field slope was less than one percent. The soil
texture that predominates was a loamy sand (Granby Loamy
Fine Sand, Figure 4.2). A loamy sand ridge ran across the
field in a north-south direction, located two thirds of the
way upslope from the east boundary. The field had two
small depressions which would pond water when the
water table was raised too high by subsurface irrigation.
The field dimensions were approximately 402 m wide and 402

m long.

51

Hmom EmoH upcmm .4 comm mo Emnmmmp omumecum

H.e mcsmfim

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_ Ill. 52. (VF in
so fl 2:: we; 2: com
sz 22.3625 _ . \ _
. . S
. _ m 9:28: 000
_ mmeom :5... u “.95 SE: 6
.l. .I I l l .l I.
_
. <mm< eme<uamma
(\vi . - -
(t, .
is _ T. s
_ m.
_ W I: mHi!
_ ms<¢mh<s
/ j _ m 3:
II _
WUQHZ ./ 1T.
_ N /
_ II
<52 / 3d:
_ _ .17 uz_¢o»_zoz
894685-202 _ 3,233 2: m: x .222:
_ . 30000
D. . z 3 4 c
. a
mDOOS -
0 Q _
_

 

 

 

 

 

4 comm co modxu Hmom no uflbmsmcuw

N.e wczmmm

 

 

 

 

 

 

 

 

 

 

 

TI
_ _ >
u _ mas
. maea>z<o
m. _ _
>mz<mu _ .
" mmaom ream _
Flillll
ma
grillilr, (till/Illliiillllr " smz<mo
\ei _ (ail
f — \T é
7. _
5 mHH . .lV
maaa>x<o _ o
. z
_ m
4mm< omeaoamma-zoz . mama ame<oamma
_
_
«N _
azzamou _ i
_
D _
_ 40¢
mecca _ omoaemme
o 0 . I.
f
A“ E HON E HON ”a

 

 

 

 

 

 

53

The layout of the subsurface irrigation and drainage
system ran east-west with laterals parallel to the major
slope of the field (Figure 4.1). Alternate laterals were of
old clay drain tile, 100 mm in diameter, installed many
years before at 32 m spacings. Prior to the study, the
spacing between the clay tile was split with 76 mm (3 inch)
pin—hole corrugated plastic tubing. Tile were placed
approximately 0.8 m deep on a 0.1 percent grade. Tile
lines were run into the loamy sand ridge and terminated.
Behind the ridge was a smaller drainage system of 76 mm (3
inch) corrugated plastic tubing spaced at 10 m. This small
drainage system was serviced by a 150 mm main which was
connected to the 200 mm field main. The drainage main ran
the length of the subsurface irrigated area on the east
boundary and was connected to the water level control
structure which was 1.2 m in diameter. The total variation
in field surface elevation between the monitoring locations
was 0.3 meters (except on the narrow ridge which was
higher). The water table elevation for the entire field
was controlled by the one structure.

There was no visible surface runoff during the growing
season indicating all excess soil water left the site
through the tile system. The fertilizer, herbicide, plant
variety, and plant population were the same for the
irrigated and non-irrigated sites. A yield check was taken
well away from the edge of the irrigated site to avoid

having lateral seepage affect the non-irrigated yield.

54

Yield checks were conducted according to the method
established by the National Corn Growers Association.

The Granby loamy fine sand is a mollisol, with the
subgroup being a typic haplaquoll, a sandy mixed mesic
soil. Small areas of Tedrow and Oakville loamy fine sand
were also present having ochric epipedons with no
subsurface horizons. The Tedrow was classified as a typic
udipsamment and the Oakville was classified as a typic
haplaquent. Both were sandy, mixed mesic soils. A
Thetford loamy sand existed in the irrigated part of the
field. It is classified as an alfisol, with the subgroup
being a psammaquentic hapludalf, a sandy, mixed mesic soil
and poorly drained. The Granby was a poorly drained
soil, while the Tedrow was somewhat poorly drained
and the Oakville was a well drained soil. A water table
was established under the Oakville soil at a depth greater
than 1.5 m near a ridge (Figure 4.1) and 0.5 m near the
farm house. During subsurface irrigation, wet sand could
be pulled up with a 1.5 m auger.

Borings were taken with a hand auger to a depth of 1.5
m. No clay barrier was encountered. Either a clay barrier
existed at some depth to perch a water table or else a
naturally high water table caused the excessive soil
moisture condition for which artificial drainage was
needed.

Water for Site A was pumped from the Raisin River. A

150 mm buried aluminum irrigation main ran 2439 meters from

55

the river to the field. This main line was shared with a
sprinkler irrigation system used by a neighbor to irrigate
an established apple orchard. The operating head on the
irrigation main varied according to the choice of pumps the
farmer used (the farmer had a choice of three different
pumps), the amount of water needed for either the
subsurface irrigation system or the overhead sprinkler
system and the pump rpm.

Prior to irrigation a fixed height overflow pipe was
installed and bolted on the end of the tile outlet in the
water level control structure. Application of irrigation
water was initiated by raising the water level in the
structure by pumping water into it. The water level in the
control structure was periodically brought up to a constant
elevation and held for a short time. Water moved through
buried tile lines into the soil to raise the water table.
Pumping was not continuous and the water table was allowed
to drift downward as plants used moisture and other
possible losses occurred. The crop grown on the field

during the 1983 study year was corn, with tomatoes the year

before.

2. Field Site B

Site B was owned by J. Ellenbaum and was located three
kilometers south of Pigeon, Michigan. The legal
description of the site was the NE 1/4 of the SE 1/4 of

section 28, R.1OE. T.16N, Huron county (USDA, Soil Survey,

56

Huron County, 1980). The field drained into the Pigeon
River via the West Branch Drain, which in turn drained into
the Saginaw Bay. The field sloped to the northwest at less
than 1% slope. The field was approximately 402 meters wide
by 402 meters long (16.2 ha), with 15 ha (37 acres)

tilled (Figure 4.3).

The layout of the subsurface irrigation and drainage
system laterals was a north-south gridiron (Figure 4.3)
with 20 meter (66 ft) spacing and tile lines diagonal to
the slope. The drainage system was installed in the 1950's
and consisted of 100 mm diameter clay tile laterals, with
the tile depth 0.6 meters at the north end and 0.7 meters
at the south end.

The field was divided into three zones for water table
elevation control, each having about 457 mm (1.5 feet)
change in the soil surface elevation. Water level control
structures were spaced at selected elevation increments
along the tile main. An overflow pipe in the control
structures allowed excess water to move from the higher
elevation zone to the next lower zone. This theoretically
allowed control of the water table height over the entire
field in three increments. The tile main sloped toward the
west which allowed water to be put in at the east end of the
main. Zone 3 had a control structure with a valve attached
to the outlet and a valve attached to the tile main on the
west side of the structure. The valve attached to the west

tile main could be shut off to prevent water from entering

57

Hwom smog >m~u

.m comm Lo uwumsmcum

m.e mesowm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l s New «i
b
_ _
_
_ _ _
_ _ Aunvnve _
_ 11 (T _
_ 17 (v _
m _ _ riomo _
H QOV .
a _ ////Lm _ meae
m mzou - _
m omeaonmauzoz _ a mzou _ N mzom _ m mzoa
a _ _
0
V
a _ _ _
22¢: mass _ . _
magma _
_ 0220.220: III 44.00 0 my _
_ 44:24: >maozaom
_ _ \ mzom
mmmaeuemem _ .
somzou am>mq _
_ _LP\K\. mme<3 xrsJ

 

 

 

 

 

2H4: mama. FMJEDO llil'l _

 

 

E mow

 

 

 

 

EDQZH mm9<3

 

mUEHQ

 

58

zone 3 or left open to irrigate zone 2. In this
experiment, the water level in zone 3 was set lower than
zone 2, from which it received its irrigation water. A
second valve at the outlet of zone 3 controlled the water
from being lost. Surface runoff from excess rain was
directed into the West Branch Drain.

The soils (Figure 4.4) were Kilmangh clay loam and
Shebeon clay loam (Michigan Soil Association group 74).
These soils are Alfisols. The subgroup for the Shebeon
was an aeric ochraqualf, clay loam mesic, and the
Kilmangh was an aeric haplaquept, clay loam mesic. The
Shebeon and Kilmanngh loam soils were classified as
poorly drained. Borings were taken with a hand auger. A
very dense clay barrier existed at approximately 1.2 m
below the surface. Artificial drainage was needed to
remove excessive soil moisture so that early spring
planting and harvesting were possible. Without artifical
drainage, a perched water table condition would exist.

Application of irrigation water was initiated by
running water into a riser pipe connected to the tile main
at the upper end of the field. The water then moved
through buried tile laterals and out into the soil profile.
Water moved to lower zones by overflowing control
structures along the tile main. Water table elevations
were stair-stepped down to follow the change in the soil
surface elevation. Excess water was allowed to overflow

the last water level control structure and flow into a

S9

 

UW>HZ>0m UHHDI

g

 

mxmmmoz FO>Z
u».

 

  
 
  

ZHFZ>Z>Qx PO>Z
m

 

It

 

 

 

 

 

 

TL+

mmmwmoz FO>Z
u;

 

mwmc1m 5.:

mnrmamnwn 0m MOWH n<pm o: anm mu

 

 

60

drainage ditch. The water supply came from a 127 mm
diameter well with a 100 mm diameter submersible pump. The
irrigation water was pumped with a three phase, 11.2 kw
submersible pump. Well depth was approximately 85 m. The
capacity of the well was approximately 380 liters per
minute (100 gals/min). The well was located in the
northeast corner of the field.

A non-irrigated plot was set aside along the east side
of the field to compare irrigated to non-irrigated yields
(Figure 4.5). This was accomplished by shutting off a
section of the tile main line, thus allowing no water to
flow into the non-irrigated field plot via the tile
laterals when subsurface irrigation was begun. No
monitoring of the water table was done on the non—irrigated
plot. Plant population, plant variety, herbicide program,
planting date, soil texture and the fertilizer program were
the same for the irrigated and non-irrigated plots. Yield
checks were taken in the non-irrigated plot well away from
the irrigated plot to avoid having any lateral water
movement influencing the yield check. The crop grown in
the site was sugarbeets with 760 mm (30 inch) row spacing.
A population count of growing plants was taken in both the
irrigated and non-irrigated plots approximately six weeks

after emergence.

Ao_mUm ob accv
m mu_m co mbo_a poumo>cmc no co_umu04 m.: mt:m_d

 

4<mwb<q MJHE ll)

 

_——-—

 

 

 

 

 

 

 

 

 

 

 

 

61

1) (Y
\Y

\A

 

 

a
soda
.. a H m m N H H
5.2055 202 .55 .515 .5151 sci 3:25:
033m:

 

 

 

 

 

 

 

 

 

62

3. Field Site C

Two field sites were selected on the B. Singer farm.
Site C was located approximately two kilometers from
Saginaw Bay and three kilometers southwest of Sebewaing,
Michigan. Its legal description was the SW 1/4 of the SE
1/4 of section 28, R. 8E. T. 15N (USDA, Soil Survey,
Tuscola County, 1984). A schematic sketch of the field is
shown in Figure 4.6.

The soil texture that predominates is a sandy clay
loam named Wisner loam. Wisner loam is a typic
glossaqualf, with a texture finer than loamy fine sand. It
is an Alfisol with an aquic moisture regime. Gleying is
very common in this soil, indicating that the soil has been
saturated over an extended period of time. The field
required only one water table control zone. Dimensions
were approximately 402 m wide by 402 m long, having a total
area of 16.2 ha (40 ac), with approximately 14.2 ha (35 ac)
tiled.

Because of the proximity to Saginaw Bay, bay water
backed up in the drainage ditches and a naturally high water
table existed. The bay maintained approximately one meter
or more of water depth in the drainage ditches year around
during this study. The top width and total depth of the
ditch along the south side was six meters and approximately
four meters respectively. Drainage water from the field
was removed by an electrically powered vertical shaft

centrifugal pump (Figure 4.7) in a 3.6 m deep sump which

 

63

.smoH xmau xpcmm .U mufim no Emummwp umbmsmzum m.¢ musmmm

 

 

 

 

 

emqeao.mll random mme<3 a :UBHE m0<zH<mo
040m mHzooa
90392004... $5.53um aomezou .52: mmefis
\III'I.

ZOHE<UOJ mZDm

 

I
\

d<mm9<q mdmfi EE NOH

.——

 

 

 

 

ZOHPUMW

20H940Hmmuizoz

“
_
_
_
_
_
_
_
_
_

 

P\\ .thPDU
. 2H4: as now

macaw

 

 

 

/l 2H4: ma~9 55 now

 

mmmamoumm h
mm»? mzm>m9m

Pozuu<mm
ITIIIVI,
S OH

 

 

 

 

 

 

 

 

 

 

 

—_————

 

 

mudmz 20 H 9<>mmmm0 1:524:

 

é
e _

 

 

T
040 A Boos
4

 

 

 

 

 

 

64

.u mbmm .dsucm need can muzuuzcbm
Hotbcou Hm>m~ cmums mo umumEmcum n.e mnzmmm

 

>Jaaaw mmk<3
Dz< :UEHD mU<ZH<mO

mx<FZH mm9<3llvi IFmJEDO

 

 

 

 

 

0mm 0401

— d— - - —
d*
—_ 0—— — — --

PDQ mm9<3
A m0<2~<me

 

mom<xUmHQ QZDQ
mqm<>02

m>4<> _

mmemzfo/ _

mmemz m4 /
draw

QZDQ Ema;

wm>J<>

 

(IIL\ int: zz<z maze

EE mom

ZH<Z de9 EE mom

65

served both as an irrigation and drainage lift pump. The
pump had a variable speed belt-pulley that the
farmer/manager could adjust to increase or decrease the
amount of water supplied to the field. Irrigation was
initiated by opening the irrigation intake pipe from the
ditch to the pump sump, closing the tile outlet valve
directing the flow from the pump into water level control
structure and opening one or both valves on the tile mains
extending out into the field (Figure 4.7). The valves were
rubber—lined metal plates that closed over the tile main
ends. The tile mains had to be either totally open or
totally closed. The farmer determined when irrigation
water for the crop was needed based on his understanding of
the crop water use and soil moisture conditions.

Irrigation water was brought from the bottom of the
drainage ditch to the pump sump via a 254 mm (10 inch)
conduit. Each of the two 203 mm (8 inch) tile mains
transported water back to about one half of the field. The
water then moved through buried tile laterals into the
water table. The tile laterals themselves were alternating
clay tile and perforated corrugated plastic pipe, 100 mm in
diameter and spaced 10 m (33 feet) apart. The tile depth
was approximately 0.66 m at the north end and 1.0 meter at
the south (outlet) end. The subsurface irrigation-drainage
system laterals ran in a north to south direction on a
gridiron layout (Figure 4.6).

To create a non-irrigated zone, the upper end (Figure

66

4.6) of one tile main was interrupted and capped off. A
380 mm sump was installed on the non-irrigated side of the
tile main just above the capped tile main. A submersible
pump was then installed in the sump to remove all drainage
water coming from the tile laterals in the non-irrigated
zone. This resulted in isolating the non-irrigated zone as
a separate area with subsurface drainage only.

The field was divided and seeded at two different
rates with emergence counts taken to determine actual plant
populations. Figure 4.8 shows that the irrigated area was
split in half, with the west half having a higher
population (66,700 plants/ha) than the east half (63,400
plants/ha). The population for the non-irrigated area was
63,400 plants/ha, the same as the lower population of the
irrigated area.

The vertical shaft centrifugal pump was belted to a
3.75 kilowatt electric motor. The pulley on the electric
motor was adjustable to achieve different pump speeds to
increase or decrease the pumping volume. A timer was
connected to the pump to record the amount of actual
running time. The instantaneous pumping rate was
measured periodically using a catch bucket and stop watch

to calculate total discharge.

4. Field Site D

A fourth monitoring site (site D) was also located on

the B. Singer farm located approximately three and one half

67

U comm co mxumzu pfimm> Lou mboda Umbmm>ncz

m.e mczoam

 

 

 

 

 

 

 

 

 

.lll.DME<DHmmHIZOZIY

_ omemm>m<m

_ _
s?

_
_
s. _

 

 

 

L.

*l

mEOJm
zUmIU
Emm>m<m

 

 

I1

_ _
_ _ q _ _ _ _ _
_ _ _ _ _ _
_me mm.o_ _me ms.s_ _ m; mm.s_
mmaeuemem
_ _ _ _ somezou am>ma «macs _ _
_ _ _ _ _ _
_ _ g _ _ l _ _
x (V k_\
_ _ _ _ _ _
_ _ _ _ _ _
_ _ _ _ _ _
_ _ . _ . zoseaulmml (w _ l
_ _
_ _
_ma soaa_ Na soda as Foam
_ _
_ .

 

 

68

kilometers southwest of Sebewaing, Michigan, in Tuscola
County (Figure 4.9).

The legal description was the NE 1/4 of the SW 1/4 of
section 32, R. 8E. T. lSN. (USDA, Soil Survey, Tuscola
County, 1984).

The soil texture that predominated is a Wisner loam.
Because of the seasonally high water table, gleying as well
as mottling were present. The clay barrier is
approximately two meters from the surface.

The proximity of the Saginaw Bay provided a year
around supply of surface irrigation water as for Site C.
Site D was also very flat, with the total area in one
water table elevation control zone. Tile laterals were 100
mm diameter, concrete tile spaced 20 m (66 feet)
apart. The gridiron tile system was approximately
one meter deep with laterals laid out in a north to south
direction. Drainage water was run into a three meter deep
sump, 1.22 m in diameter. A 0.56 kilowatt, submersible
pump was utilized during drainage to lift the water to a
second vertical pipe which had a 203 mm (8 inch) conduit
running from the bottom of it under a road into the
bottom of a drainage ditch (Figure 4.9).

Water was delivered to the artificial drainage system
for irrigation by reversing the process. Water flowed from
the ditch through the 203 mm conduit into the second
sump. The pump was shifted from the first sump to the

second, and the water was lifted and deposited into the sump

69

EmoH xmau accmm .0 002m uo Emnmmwp owumsmcum m.e enemas

 

 

ZOHB<DHmmH mom BUDDZOU a

LI 2.3.2.00 Samem was; 024 20.30 m0<22<20

 

 

>43 Q<Om

 

 

NIVN TILL

 

 

T203504 axed
mmmneusmem 4029200
.L\ amt/ma 2543

 

 

ma<mm9<d mane mfimmuzou

 

 

magma
Ill 20.520: 2022 Li.

(9000

0 00C?

 

 

I

 

 

X

 

UZHU<mw mmfimz om

 

 

 

Alli! mmoqm

 

 

 

 

 

 

 

 

 

 

70

which was connected to the tile main. The water then moved
through buried tile laterals and into the soil.

The water level was controlled by turning the pump on
and off with no provision for varying flow rate. A timer
was connected to the pump to record the amount of running
time on the pump. Water volume being pumped was measured
periodically using a bucket and stop watch procedure to

calculate the total discharge rate.

C. Water Table Monitoring

The movement of the water table under subsurface
irrigation and drainage conditions was measured using a
series of perforated vertical monitoring tubes placed in
the soil profile. Tubes were placed at each end of two
adjacent tile lines as shown in Figures 4.1, 4.3, 4.6 and
4.9. Two different types of tubes were used. One type was
13 mm PVC tubing approximately 1.2 m deep. The water
level was recorded manually in the wells on all four field
sites using a blow tube of flexible rubber (Figure 4.10).
The second set of wells were 76 mm diameter PVC
tubing placed approximately 1.5 meters deep. The water
level in each large diameter observation well was
continuously recorded with Stevens Type F water stage
recorders.

Traces of water level vs time were obtained with
Steven Type F water level recorders (Figure 5.11) for the

1983 growing season on Sites A and C. Manually recorded

7l

.xflamscms mcudmp wanna qum3
203.205 00 tom: mbsbusofin m “0 cowumnbmszHou€ mesa;

hv.
O
o All) .155 mmemmsqom
O
0me<20m2mm o
mmae 0220.520: 0
BE as NH I! all g All 4562.3 00:.
O

wmqmmbm a“?

. 3 ii mam<9 mmp<3

mfimmo QmmDm<mZ

 

/AvayRXAAMAaQXWAMAOAAAOASAOWANKOANA”AdAnxvv \vva/0XVY&X&AMQVOvv¢oymvoyuvvxvoxquwvm

mu<220m 020m
20He<>mqm 23022.: .
mass 304m .
..IIIIIII 224
as m

 

 

 

72

.co_um>o_o o_nmu teams mc_mcm;u mc_czmmos to»
20020002 mmmbm Loam: mco>oum 0.00304 to u_umsocum p_.: mc:m_d

KNKNKKXVV

mesa ome<moamma
u>a as on It

 

.N%?§¢§XW%&%

mm Hmm<m mDO H >mmmZH

‘nlllul Emu H m3 mmEZDOU

llll'llll

man—mama. mm9<3

‘l mfimmo Dmmbmfimz

\\>//\\>// \\\/o\\\a\\\u/\b\4\\\a§éwu

 

 

 

 

 

 

mu<mm3m a How

 

 

 

 

 

‘Illmmomoumm MOS—.m mm9<3

 

 

zm>m9m QdOmDmd

 

73

water table records were obtained on all Sites A, B, C and
D. The water levels at each site were managed by the
farmer according to his normal management practices. Where
continuous records using water level recorders were
obtained, events such as the rise in the water table versus
the time when the farmer initiated subsurface irrigation
were recorded along with diurnal variations over and
between tile laterals. To characterize the response of the
system to irrigation and drainage, events having a rising
water table or a descending water table were analyzed from
the water level recorder charts. The response of the water
table to drainage conditions was recorded when irrigation

water was allowed to drain out of the tile system.

D. Saturated Hydraulic Conductivity

The saturated hydraulic conductivity (K) was determined
using a Velocity Head Permeameter developed by Dr. G. E.
Merva, Michigan State University. Results are presented in
Table 1. Depth of the measurements ranged from 15 to 50 cm
in both the horizontal and the vertical direction for Sites
B, C and D. A depth greater than 50 cm should have been
measured for saturated hydraulic conductivity. However,
this was not accomplished due to lack of foresight on the
researcher's part. A depth of about 70 to 80 cm would have
given K values closer to the depth of the tile lateral. At
Site A, the sandy soil profile slumped into the borehole so

the velocity head permeameter could not be used, and

74

hydraulic conductivity was taken from the estimate in the

USDA, Soil Survey Book, Monroe County, 1984.

E. Crop Yields and Planting Inputs

Final yields were reported in Table 2 for each site.
Corn yields were measured using methods established by the
National Corn Growers Association on Sites A, C and D and
expressed as yield at 15.5 percent moisture. Yields for
sugarbeets were measured on Site B. A tare was subtracted
for the soil and residue removed from the sugarbeets as
they were unloaded at the processing plant. A second tare
would also account for the weight of the remaining soil
left on the sugarbeet during storage which was estimated at
5% of the total weight and subtracted from the yield.
Plant populations were determined through an emergence
count (Table 3). Data on crop inputs including corn
varieties, fertilizer and herbicides were obtained from the

cooperating farmers (Table 4).

F. Available Soil Nutrients

Soil fertility influences the performance of a growing
plant. Soil samples were taken from each of the fields
used in the experiment and sent to the soils lab at
Michigan State University. Phosphate, potassium and soil

Ph were measured and are recorded in Table 7.

75

G. Evapotranspiration

Evapotranspiration (ET) was estimated for corn using a
modified version of the Penman equation in a spread sheet
format as described by Vitosh (1984). The program utilizes
minimum and maximum temperatures during each growing day,
the geographic location within the state of Michigan, and
the crop coefficient (KC) which relates evapotranspiration
to stage of crop growth for the estimation of ET. The
stage of crop growth is calculated from the emergence date
and genetic growing period of the corn variety.

By inserting the appropriate coefficients into the
spread sheet, accurate calculations of the crop evapotran-
spiration (ET) at each stage of growth for a specific
crop, location, date, and mean daily temperature are
possible. Once the spread sheet is set up for crop and
local conditions, entries of any deviation in daily
temperature from the long term mean and any rainfall or
irrigation amount on the date it occurs can be made. Root
depth, profile moisture content and the amount of
irrigation that may be added on any day are automatically
calculated. A graphic display of the daily moisture content
in the soil profile is shown. The computer program begins
reducing evapotranspiration rates when soil moisture
content is less than 50% of the available moisture.

The calculations for evapotranspiration were begun

after a major rainfall event. Because soil moisture was

76

assumed to be at field capacity following the rain,
potential evapotranspiration was also assumed. Because
subsurface irrigation seeks to establish a water table in
or near the root zone, it was assumed that after irrigation
commenced, the available water supply to the plant would
always be greater than 50% available soil moisture, and ET
would never be limited. Therefore, the ET rate that was
potentially possible for the crop being grown was

calculated.

H. Precipitation

Precipitation amounts and date of rainfall events were
obtained from two sources. The first source was the field
location where local rainfall data was recorded by the
farmer through the use of a rain gauge set up above the
crop canopy. The farmer was to record any precipitation
each morning in a record book for his field location. The
second source was the nearest weather station. The
rainfall amounts and their corresponding dates were
compared. A check on a rainfall event was made from the
water stage recorders for Site C. When the water stage
recorders showed a rapid water table rise and this
coincided with the weather station records for rainfall
while the farmer's recorder revealed no rainfall for that
date, rainfall was assumed and the weather station record
was utilized. However, every effort was made to use the

farmer's rainfall record because of the possibility of

77

local variation in rainfall. These values were compared
with local rainfall data recorded from other local weather
stations like airports.

The amount of drainage water which was removed from the
system was not measured due to the nature of the subsurface
irrigation systems designed and installed by the farmers.
The farmers would drain the systems whenever they felt
excessively high water table conditions would result from
rainfall amounts. The water stage recorders measured water
table elevation as it changed over time, including ET
induced fluctuations and changes due to irrigation and
drainage.

Sites A and B did not result in a total water volume
being recorded. The reason was that the farmers at each
site took liberties to vary the pumping rate as crop and
precipitation dictated and failed to record changes in pump
output operating pressure, pump rpm, and the length of time

the pumps were operated.

V. RESULTS AND DISCUSSION

Water table elevations at four sites were monitored
during subirrigation. At two sites, Sites A and C, both
continuous and discrete data were collected. Leupold Stevens
Type F water stage recorders were used to automatically
monitor the water table elevations near and midway between
tile lines at selected locations in each field (Figure 4.10).
At Sites B and D the water table was monitored manually
utilizing monitoring tubes at locations interspaced between
parallel tile laterals. A blow tube was used to determine

water table depth from the datum selected for each site

(Figure 4.11).

A. Site A

The results of the field evaluation to determine the
response of the water table at Site A yielded very little
reliable data. The data was not adequate because of the
following reasons: 1) the time of pump start-up,
discontinuation of irrigation, and other information such as
pump speed and pressure were not recorded; 2) the soil at the
site, a loamy fine sand, moved into the tubes housing the
float and counter weight, causing them to stick, resulting in

no data; 3) measurements made by two water stage recorders

78

79
using 30 day timing clocks were not consistent with each other
on water table fluctuations from a predetermined datum; and 4)
cables connected to a number of floats and counter weights
jumped the pulleys driving the drum which contained the charts

for recording water table change.

B. Site B

The results of the field evaluations to determine the
response of the water table at Site B are recorded in Figures
5.1 through 5.4. All measurements were made manually. The
positions of the monitoring tubes are noted in Figure 5.1
through 5.4 by the vertical arrows at the top of the figures.
The tubes were spaced at five meter intervals between two
parallel tile laterals, with a tube near each lateral. The
tile lateral spacing was 20 meters. A clay barrier existed at
approximately 100 cm or greater below the soil surface. A
datum, which was the top of the supply well outlet pipe, was
used as a common reference. The bottoms of the monitoring
tubes were approximately 77 cm and 100 cm below the soil
surface at the north and south end respectively. The tile
lateral depth varied from 42 to 47 cm below the soil surface
at the north end of the field and approximately 83 cm below
the soil surface at the south end of the field.

Figures 5.1 and 5.2 show the Site B water table
elevations at the north end of the field, close to the water

supply, for different observation days. Figures 5.3 and 5.4

80

.cOmmom >_Lmo .m mu_m
.cco gyro: .moump u>_w co mco_um>o_o 0_nmu c0003 ..m ot:m_d

A8395 6.32 0.3 E0: 8:0me

 

om mp op m o
r n n F Li — L )- - Ll L b P i? P h b b L L _
mlm x
ale 0 Aobogxoaaaov Cotton 20.0
min a llllllllllllllllllllllllllllllllllllllll
mmln o
owl. u
\monau octogcoE Co Eaton
ml llllll a llllll In) llllll .I. llllll ID

® .832 22 O

X

X

0| i \oootam :3

p p p >
\\ >

c0332 33 mc_._o:coE\\
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Egon

”m 35 co use 552 625004

0.:
On..—
our
Ofi F
00*
OO
cm
on
cm
on
O...
on
ON
or

O

(me) 1110le molaq souoisgg

81

.200000 000— .m 00_m
.020 £000: .m0um0 0>_0 co mco_um>0_0 0_000 c0003 ~.m 0t:m_d

A00000EV .0020. 0.: E0: 00:305
on m. or m
Elm
5(0

0

t L
b
m
calm m
e
w

nouogxotaaov Cotton >20 Won?

hlm

0003 9:33:05 00 Eozon
IIIIIII iIIIIIIUr

)4".

'

 

002.50 zoml)

.-
.-

5580. 003 octotcoE

m» D4 .QKII

P—VjilYYIUIjWWT

OO
,...

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 8280

5 03m Co 0:0 5.32 "cos—003

(we) (uniop Moleq souoisgg

82

.2Omm0m 2.200 .m 00_m
.020 22:00 .00200 0>_0 mo mco_um>0_0 0.202 20203 m.m 023022

$202025 62022 2: 22022 0020205

 

 

 

 

ON m_ O. n O
Vii-WNW. Xlr L L.xl||»l.-..--IILII(IILrIo .IIrI-IILII;IILZI-.IL: IL- (IL.-.- -Lfililr: flue--- IL :IL. .- IL “.00.—
elm 0 02022202003 2022202 >20 Wom—
nlm a IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII A
main 0 wom—
omle n m
0:
\0023 0220:2922 20 E0302 n
m. iiiiiiiii g iiiiiiiii EIIIIIIII iiiii (EIIIIIIIIIIIIIIIIDVOO—
a
mom
mom
0 62022 0_:.IOWO\.
m8
2
X X won
moi.
MW MW won
mom
\0002230 :00 mo—
uI-.I.I.I..|..I)iIIIiIIiIallaIIII-Iill..II...II.I.II.I:I..:IIII...|I.I..III:..|.I...HuLI.iHuuHHnHHHIIHiHHIIIHlIsifiHMJMOUI WO
2 . 21/ 2 2

203000. 023 0220:2022 (II-l.

"m 02m 20 020 530m "2020000

0;) M0;eq 300:);ng

‘
If

(0.0) um

83

.200000 000. .m 02.m
.020 20:00 .00000 0>.0 co mco.u0>0.0 0.200 20003 :.m 0200.2

320.025 620.2 0:. E022. 0020.05

 

ON 0. O. m O
r t t i r 2 t L t i 2 t L I F- 2 . t t i 2
n.Im b
.nlm m 02022082003 20.2202 220
ONIO x IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
mlm e
hlm n

0023 02.23.2022 .0 E0302 J

 

x m
w...
M >
M

“000.230 :00
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII HHHHHHHfiHHHHHHHHHHHHMEmmfl
u p p ...l p »

22.002 023 mc.2o._co_.c\‘.
um 02% .0 020 £30m ”20.20004

H00.
Won.
.02
m

00:

moo.

OOOOOOOO
NW‘TIDLDNCDO'J

O

 

[1111171111111111111ITIIYITITWII‘IIIIIUrITIIIIIT

O

wniop Moleq aouoislg

\

A
v,
1'

(0.;

84
present the elevation of the water table at the south end of
the field, away from the water supply but located on the same
laterals as monitored at the north end. Figure 4.3 shows the
location in the field where these measurements were taken.
Data was recorded for 56 days during irrigation, from 7/20
through 9/13. Prior to the beginning of monitoring, the
system was in the drainage mode. The water table elevation
was below the tile laterals. When subsurface irrigation was
initiated, the water level was raised in a riser pipe on the
tile main to an elevation near the soil surface. The farmer
did not record the initial start-up time and date of
irrigation nor when the pump was shut off. However,
irrigation was initiated either before or on 7/28. Irrigation
water input was operated in an on-off sequence prior to 8/8,
then continuously from 8/9 through 9/13.

The water table elevation at the north end (Figure 5.1)
reveal a rise near the tile lateral on 7/28 and by 8/3 was on
the soil surface. It was held there through 8/4, allowed to
decrease below the surface on the 5th, and brought back to the
soil surface until August 7 when the pump was shut off. The
water table then decreased below the tile laterals by August 8
(Figure 5.2). From August 3 - 7, water above the tile
laterals was visibly noted. Water was puddled on the soil
surface over the tile lines at the north end of the site.

At the south end (Figure 5.3), the water table showed a
lag in the time of rise. The water table did not show a

response on July 28 as the north end had done. However, by

85
the time of the next observation on August 3, the water table
had risen to approximately 27 cm below the soil surface,
similar to the rise in height at the north end. It could also
be noted that as one proceeded south from the north end, the
puddled water ceased and only capillary soil moisture appeared
over the tile laterals. This can be understood by analyzing
Figures 5.1 and 5.4. The water table at the south end never
came to the soil surface. The soil sloped upward, making the
water table approximately 27 cm below the surface but at the
same elevation as the north end. Capillarity moved moisture
upward 27 cm from the water table to the soil surface.

The observations on August 8 revealed the water table
dropped below the tile laterals at the north end of the field
and dropped to just above the tile laterals at the south end.
This was a decline of approximately 48 to 58 cm at the north
end and 40 cm at the south end in about a 24 hour period. At
this point the water supply had been turned off, and water
flowed away from over the tile laterals to cause the decline.
Reasons for the rapid decline may have been that water was
being removed by evapotranspiration over and near the tile
lines by the crop and/or water was lost as leakage near the
outlet pipe.

The exact time when the irrigation water was turned back
on was not recorded. The exact date of the beginning of the
water table rise before August 20th at the north (Figure 5.2)
and south ends of the field (Figure 5.4) was not known.

However, the water table over the tile laterals was held

86
between 0 to 15 cm below the soil surface at the north end and
30 to 40 cm below the soil surface at the south end until
September 13th.

Figures 5.5 and 5.6 illustrate the difference in water
table elevation at the two extreme ends of the field, but over
the same tile lateral. The break in the water table
elevations, as shown in the figures, represents a time
period during which the rate of rise of the water table was
unknown. The figures allow some insight into the effect of
distance from the water source and friction in the tile
lateral on the loss of water table height.

Figure 5.5, representing the west tile lateral of the
area of the field being monitored, shows a fairly uniform
water table response at both ends of the field. There was a
delay in the initial water table response at the south end,
but by 8/3 a uniform response was seen.

Figure 5.6, representing the east tile lateral of the
area of the field being monitored, showed less agreement in
responses at both ends of the field. This may have been a
measurement error or head loss due to friction in the pipe.

In checking the decrease in the water table elevation
over the length of the tile lateral as a result of friction,
the Manning formula for open channel flow in field drainage
tile is often used (Schwab, 1966). The Manning formula is
normally applied to tile drainage because very seldom are the
tiles flowing full of water. With subsurface irrigation,

however, the field tiles are filled with water and pipe flow

87
condition occurs rather than open channel flow. The Hazen -
Williams formula as outlined by Finkel (1982), therefore,
applies. The Hazen - Williams formula is historically a
proven formula and has been used for many years to describe

head loss during pipe flow.

The Hazen - Williams formula is:

h1 = (Q/C)x * KlL/Dm ....................... 5.1

where h1 = head loss (m) through L distance; L = length of

pipe (m); Q : flow rate (m3/sec); D = internal diameter of the

pipe (m); K1 10.59 (SI units); 0 = roughness coefficient; m

= 4.87 and x 1.87.

Q was calculated by taking the rate of water input
(0.0063 m3/sec) and dividing it by the number of tile laterals
in the irrigated portion of the field.

Because it was difficult to calculate the head loss in a
concrete tile lateral with water being removed through
perforations or joints, the exact amount of water removed as
it flowed down the tile lateral was unknown. Therefore, it
was assumed that the total volume of water entering the tile
lateral from the tile main was transported the full length of
the tile. Maximum head loss could then be calculated assuming
no water loss. The following parameter values used were L =
372 m, Q = 0.0004 m3/sec, D = 0.1 m, K1 = 10.59, C = 110 for

old concrete tile (Finkel, 1982), m = 4.27 and x : 1.87. The

88

m 02m .20: 0E. .20 00:0 £02 20 00202008 0::
03 2002, 0S. 20>0 c220>0_0 00203 mh 0.39m

 

A0E3 0200
mr\m m\m n\m mm\m ¢m\0 m_\0 ¢P\m m\m ¢\m on\n mm\n om\e
{ h r Ir b F b L Lb P L: L
20032300 I
20022220: 1... 3323303 22:02 >20

0023 octotcoE *0 E0302

 

 

o:
of
03
o:
02
om
00.
E.
8
on
c...
on
cm
9
o

(000) wnqop Mogeq 0:300;ng

89

m 0:m .20: 022 20 00:0 £02 20 0020200,: 0::
03 2000 05 00>0 c220>0_0 20203 02 0.59.2

 

A0Esv 020.0
n_\m m\m n\m m~\0 ¢m\m m_\m ¢_\0 m\m ¢\0 on\n mm\n o~\n
p — L b .- + IF LP L L b L
umomrzaow III.
20005.5: ll A020E_xoeaaov 02:02 >20

 

 

HO¢F

1
1-

we?
wom—
we:
woo—
mom
mom

 

[‘TTTIIITIIUIIIIIIIIII'IIIIITTTITI'YIIVU
O O O O O O
N (’3 V“ LO CD I\

00

(we) wniop Moleq SQUDlSK]

90
result was that h1 = 0.02 m (2 cm) of head loss when Q was
transported for L distance.

In actuality, irrigation water is lost into the soil
profile as it moves down the tile lateral during subsurface
irrigation. As a result, the actual head loss due to
friction would be less than 0.02 m. Because 0.02 m is a very
small amount of head loss due to friction, it can essentially
be assumed to have no influence on the water table height over
the tile laterals.

Monitoring tubes five meters off each tile lateral
revealed no water table during 56 days of monitoring. At no
point during subsurface irrigation was a water table
established between the tile at either end of the field. The
water table came up over the tile lateral but fell off within
the first five meters from the tile indicating a large
hydraulic gradient existed in the vicinity of the tile
laterals.

Factors which affect whether lateral movement away from
the tile lateral will raise the water table are the rate of
vertical and lateral movement and the ET rate. The lateral
hydraulic conductivity and the hydraulic gradient created by
the water table over the tile laterals determine the flux away
from the tile.

Estimated evapotranspiration for sugarbeets in the month
of August varied from 3.8 mm to 5.5 mm per day and averaged
4.6 mm (0.18 inches) per day. The well providing the water

supply produced approximately 378 liters per minute (100 gpm)

91
or about 545 m3 per day. If ET averaged 4.6 mm per day and
was calculated for 13.7 ha (non-irrigated area omitted), the
total ET for the irrigated area equals approximately 600 m3
per day. The well water supply rate was below the ET
requirement, but only by 55 m3 per day (about 10 percent). A
fraction of the total possible area, an area less than five
meters on either side of each tile line received water. When
the area between the tile laterals is not considered
irrigated, the well produced more than enough water for those
areas being irrigated. Some of the excess ponded water over
the tile laterals actually ran off as surface runoff at the
north end of the field. This runoff was not measured.

The difference in the vertical and lateral movement of
the water table in the soil profile may be explained in part
by the difference between the vertical and lateral hydraulic
conductivities. The saturated hydraulic conductivity values
recorded in Table 1 show that the saturated vertical
conductivity at 30.5 cm depth, range from 2.3 to 7.7 cm/hr
compared to 0.3 cm/hr for the saturated lateral hydraulic
conductivity. Thus, the vertical conductivity was 13 to 25
times greater than the lateral conductivity. The lateral
conductivity at the 50 cm depth was also quite slow, ranging
from 0.3 to 0.4 cm/hr. Measurements using the reverse auger
hole method as described by Reeves (1982) showed a hydraulic
conductivity of 0.1 to 0.2 cm/hr for saturated horizontal
flow. Equation 3.1, Darcy’s law (Hillel, 1982), was used to

calculate the flux rate or volume of water flowing through a

92

Table 1. Saturated hydraulic conductivity utilizing
velocity head permeameter.

Field depth (cm) Lateral (cm/hr) vertical (cm/hr)
site
A —* _*
B 30.5 0.3 7.8
0.3 2 3
0.3 ~
0.3 -
50.0 0.4 -*
0.3 -
0.3 -
C 15.2 5.1 3 2
5.2 -
50.0 4.5 3.2
4.2 3.2
- 3.5
D 30.5 4.7 10.8
4.7 9.0
3.6 6.7
3.9 -
50.0 6.0 -*
7.3 -
5.6 -

* Because of sidewall slump and the existence of a water
sand environment, the velocity head permeameter and the
reverse auger hole method were not feasible.

# Vertical hydraulic conductivity was not taken at this
lower depth.

93
unit cross-sectional area per unit of time. If the soil
properties were such that the hydraulic conductivity was very
small, the volume of water (flux) moving laterally through a
cross-sectional area of soil profile would be small.

If one looks at the evapotranspiration (ET) rate in
relation to the hydraulic conductivity needed to sustain ET, a
comparison between the measured K (Kmoa) and the calculated K
(Kcal) value can be made. If the Kcal needed to sustain ET is
larger than Kmea, a water table should not have been
established at least five meters off the tile lateral.

Calculating the total evapotranspirati0n°volume (ETv)

with an ET rate of 0.0046 m/day over an area of 5 m2,

ETv

0.0046 m/day x 5 m x 1 m .......... (5.2)

ETv 0.023 m3/day/ 5 m2 area.

Calculating the lateral flux into the soil over the tile
lateral where the water came up to the soil surface, the soil
must conduct the full amount, 0.023 m3/day, through an area 1
m by 0.8 m (depth of tile below surface) to supply ET. Flux

(q) at the tile,

q : .023 m3 d ......................... (5.3)
1 m x 0.8 m
q : 0.03 m/day

94
Calculating the required hydraulic conductivity (K) to

move 0.03 m/day using Darcy’s Law:

 

Kcal = g L = 0.03 m/day * 5 m .......... (5.4)
Ali 0.8 m
Kcal : 0.2 m/day

where L = 5 m (distance to the monitor tubes), q = 0.03 m/day

and AAH

0.8 m (height of water table above tile and at the
soil surface).

Therefore, to sustain an ET rate of 0.0046 m/day, with a
lateral flux calculated at 0.03 m/day, a Kcal value of 0.2
m/day is required.

The Kmea from the velocity head permeameter averaged
0.072 m/day (0.03 cm/hr, Table 1).

The measured hydraulic conductivity value is 2.6 times
smaller than the calculated K value for the assumptions
considered. Water moving laterally could be drawn upward to
replace water used by the crop before it moves to a monitoring
site five meters from the tile lateral.

The hydraulic conductivity was small enough to explain
the fact that insufficient water moved laterally to supply
daily ET, let alone supply water to establish or maintain a
water table.

Deep seepage was considered negligible in its affect on
lateral water movement. Soil borings were obtained during the
early part of the spring when the field was too wet for field

Operations. Approximately six borings taken around the field

95
to a depth greater than 125 cm all resulted in an almost dry,
slowly permeable subsoil being encountered. Therefore, very
little deep seepage was evident.

The conclusions resulting from the monitoring of the
water table movement for the type of conditions at Site B are
as follows: 1) The hydraulic conductivity, with values
between 0.1 to 0.4 cm/hr, became the limiting factor in water
not moving fast enough laterally to establish a water table
five meters off the tile lateral; and 2) water supply was
approximately 10 percent below the capacity required to
irrigate the total area. However, it was adequate because the
area between tile laterals was not being provided with

subsurface irrigation water.

96

C. Manual Water Table Response for Site C and D

Manually determined water table elevations resulting from
a constant water level in the head control structure are given
in Figure 5.7 for Site C, and in Figure 5.8 for Site D. The
figures represent the monitoring locations furthest from the
water source and control structure. The area midway between
the tile laterals was believed to represent the position
slowest to respond to subsurface irrigation. This location was
slowest to respond because of the distance from the tile
laterals, effects of the soil properties and the influence of
evapotranspiration.

Measurements were taken manually from monitoring tubes
interspaced between tile laterals as shown in Figure 4.6 and
4.9. Sites C and D had a 10 meter and 20 meter tile spacing,
respectively. Prior to the beginning of the test, the water
table elevations were below the tile laterals for each site.
When irrigation was initiated, the water table in the control
structure was raised to a constant elevation. It was then
held at that elevation until the farmer decided to lower the
water table due to excessive irrigation water and/or rainfall.

Figure 5.7 shows the water table between laterals at the
north end of the field for several manual observations after
initiation of water input on Site C. Subsurface irrigation was
initiated on 7/18 at 0800 hours with an irrigation rate of
approximately 1000 liters per minute (265 gpm). A constant

head was maintained in the head control structure 72 cm above

97

.u 00_m .000003000 .000000 5000 000200:0.0_00000_
00020 000000 00E_0 0:o_00> 00 0co_00>0_0 0_200 00003 >.m 00:m_0

#0000080 05 E000 0000005

ON m— OP m o

— LP L P h L L L L b L L L L b L L L L L l—

?005080003 020002 >20

2882 2:0

 

 

 

mm\0
0: m¢N H0Q|||IIIIIIIIIII11 111
0: mfimn 0 n\m: ...lluXI
H KW1 Illlll 5, 11111111111 tllllllllllilllu .. :lx
0000030 __00\ \

00303000 _000:00 2

_0>0_ 00003
by Ky my my my

4/020002 023 mc_000_coEr|I|\\\‘

2.0 u 538

  

 

(UJO) wmop moleq 00001510

98

.o 00_m .003000000 .000200 5000 00020030 .0_00000_
030 000000 00E_0 0:o_00> 00 020_00>0_0 0_200 00003 w.m 00:m_0

A00000Ev 0:0 E000 0020005

l|nIIIIIle III!"
‘II‘

 

02 mm "0 l}! ./0

- \X‘i- -- I .
2 SN uzo\\.flfl\\\\\\ -i- /./

E 00... "00?... -- --m.

00300000 300200 2

\

.05. 00003

D D

220002 0230 0200220000
00000000 :00/

If
ll"
[III-ll" I'll-I'll I'll-I'll I'll
'lllll
[I'll
Ill
[III-l

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 0.01m. 00300

 

D/ D D

 

(we) wniop MOIBQ 000:);ng

99

the tile center at the position in the field where the
monitoring took place. The average depth of the monitoring
tubes was approximately 77 cm below the soil surface. The
water table prior to 7/15 was somewhere below the bottom of the
monitoring tubes. The data for 7/19 was the first measured
water table profile 30.5 hours after irrigation was initiated
and reveals a rising water table near the tile. The water
table was approximately 67 cm below the soil surface at this
time. From 7/19 to 7/28, the water table rose 38 cm in 212.5
hours (0.18 cm/hr) midway between the tile laterals at the
north end of the field. Between 7/19 and 8/3, it rose an
average of 63 cm in 355 hours (0.18 cm/hr).

Figure 5.8 shows the rise in the water table at Site D.
The supply of irrigation water was approximately 500 liters per
minute (132 gpm). A constant head was maintained 54 cm above
the tile center. This water column provided the head to move
water through the tile. The tile laterals were approximately
100 cm below the soil surface. Subsurface irrigation was
initiated on 7/16, at 1600 hours. On 7/15 the water table
profile was known to be below the tile lateral and greater than
106 cm below the soil surface, the depth of the monitoring
tubes. As a result of not knowing the exact water table depth
at initiation of irrigation, the height of the water table
change was referenced to the water table reading on 7/19. The
water table reading on 7/19 showed a concave upward shape 68
hours after initiation of irrigation, being higher over the

tile laterals than midway between. Midway between tile

100
laterals, the water table by 7/28 had moved upward uniformily.
It had progressed 11 cm from the previous monitored position
(7/19) in 214 hours (0.05 cm/hr). By 8/3 the water table rose
another 15 cm in 360 hours from 7/19 (0.04 cm/hr). The slow
rate of rise can be attributed to the lower pumping rate and a
wider tile lateral spacing.

For subsurface irrigation with a constant
evapotranspiration rate at the surface, the steady state water
table takes the general shape given in Figure 5.9. Skaggs
et.al. (1972) utilized an approximate algebraic expression
developed by Fox et.al. (1956), which utilized the
Dupuit-Forchheimer assumptions describing the water table. The
equation may be written in the following form to determine the
tile or ditch spacing required to not exceed a preset maximum
difference in the water table elevation directly over and

midway between the tile laterals, h1 - hzt
L = ( AK * (hf - hi))°-5 ............... (5.5)
u

where L = the distance between tile laterals; K = the lateral
hydraulic conductivity; u = the evapotranspiration rate; and h1
and hz = the distance of the water table above the impermeable
layer at the tile lateral and midway between the tile laterals,
respectively.

A procedure developed by Bower and van Schilgaarde (1963)
for predicting the rate of fall of the water table in tile

drained soil can also be applied to a rising water table.

101

mU<mmDm

.x\%\ 0.. \0. \% \%

®

1)

_

mmumm<m mdmdmzmmmzm

_ _ _ _

N

2

0

05000.5 0.02. In 0

- \% >0. Ex}...
0 m.

.‘L

 

0

‘—

2

2200

ZOHE<mHmmZ<mEOm<>m

H
H‘—

‘—

‘—

*—

‘—

*—

*L-

*—

:0

 

Figure 5.9 Schematic of water table profile during

subsurface irrigation.

102
Skaggs et.al. (1972) characterized the upward water movement of
the water table under subsurface irrigation conditions by
utilizing an equation derived from the methods of Bower and van
Schilfgaarde. If the water table rises uniformly without

change of shape, the flux is uniform and can be expressed by

.......................... (5.6)

(LID.
d‘ 3

u
f

in which m = the height of the water table above the center of
the tile at the midpoint between drains; t = time; u = the
instantaneous subsurface irrigation water input rate or upward
flux minus the evapotranspiration rate (units of length/time)
and f = the fillable porosity.

The fillable porosity (f) needed for Equation (5.6) was
determined according to the method outlined by Vomocil and

Black (1965):

AFPso cm : Sat.Wt - Wtso cm ................. (5.7)
Vol

where AFPso on = air fillable porosity at 60 cm of tension
(%); Sat.Wt = the saturated weight (gms); Vol = volume of the
soil core (cm3); and Wteo cm = weight (gms) at 60 cm of
tension.

Soil cores were taken on Site C and D at a depth of 30 -
38 cm and 68 - 76 cm two years after the field data on water
table were collected. The values for fillable porosity were

calulated and listed in Table 2.

I"-|---l"'|'|"|l'|l-'ll"III'llllll'|'ll||-|lll|l.-llll'-"'Il""l‘I-I-.lllIll-|"II'|'"'"|'I|",I-| II ..IIII‘.-!I"II'OIII"III ,

ll-vl|lool|ll'|.|'l"'..l|llil.CII-'IIIIU"IUIllll'llllllllllIv'llI'I'IIIIAI‘III..-"Unlv'Iull-" Ic'-"llI-|Illvlnlll'-ll-'llll .-IOltlv. t -.lytIlIIIn-Iulrlllelclrlltll II .ll..ll. !!!

 

 

 

 

o 0 n.0n0 N.NN5 o.nu n.0om 000.
v.—. O.vv 0.0 n.0n~ H.005 0.0Hm c.05m 5v¢0
N.N0 o.n 0.0 n.0n— c.0mo 0.00m n.0vm ovc0
0.00 M.N¢ 0.0 n.0n0 0.005 5.00m 0.00m enc—
n.m m.5n 0.0 n.0m— 0.505 5.0.0 0.o¢m @500
o.m n.ov 5.0 n.0w0 5.005 H.0mm r.nmm 5N0—
m 00.0 20000 10 cm - :0
0.00 a UOILO>¢
N.N0 m.mv 0.. n.000 O.N05 0.0mm w.mom we.
N.—— O.H¢ 0.0 n.0W— 0.005 0.0Nm n.50m ove—
n.o_ N.O¢ 0.. n.0n— _.NH5 n.0nm 0.05m 0v¢0
o.m o.mn 0.0 n.0n. 0.HV5 n.m¢m H.050 mnv0
0.5 «.mm 5.0 n.0n0 0.0m5 n.0om n.0mm onc—
m 00.0 24000 um IN: IN
N.O— « 0010U)¢
0HNO 0.00 n._ n.0n0 0.005 o.n0m m.nnm nun—
c 0 .nc 0.0 n.0n0 N.005 n.0Nm 0.000 ¢NNO
0.00 0.00 0.0 n.un— «.mN5 N.mnm m.N5o «NM.
0.0 n.0c 0.. n.000 c.0N5 O.nnm 0.¢Om 50m.
n.00 5.mv 0.0 n.0n0 0.0N5 N.5nm 5.n5m 00N0
m 000w madam aw NM n 00
0.0 u ~00Lm>¢
0.00 0.00 c.— n.on0 o.¢n5 ¢.v¢m «.050 00“
N.o0 o._v 5.0 n.0n0 n.oc5 o.mnm N.vom O0N0
_.N0 0.00 0.0 n.0w0 5.nM5 0.0nm 0.05m oom—
¢.o H.0m 5.. n.0n0 ¢.No5 n.00m 0.000 DON—
o.5 5.00 5.. n.0m0 0.5n5 N.mom 5.000 «ON—
0 044w sauna aw am: mm
0N0 >0000000 0x0 >0amoLoa AMEU\EOV .200 .50. a! .IO. 0: wt 0 EU 00 .100 u! Loneac
....... mummmem------s---u-nmmwmwuu---xumwwmmwmuu--uuuz 02.2 >00 co>0 .oanau co_ucu» 2000003000 0000

 

o a 0 0000 .020000 0:» 00 >00mo0oa u_aa___0

.N ofifig

104
A steady-state relationship such as Equation (5.5), can be
used in Equation (5.6) to describe the rate of rise of the
water table. Solving Equation (5.5) for u, substituting into
Equation (5.6) and integrating between to, mo, and t, m, yields

the following equation (Skaggs et.al., 1972),

Zd+y1+m

 

 

 

 

(t - to) : fL2 1n y1 - In ....... (5.8)
8K(d + y1) 2d + 11 + mo
y1 - mo
in which mo : the between lateral water table elevation at t :

to measured from the center of the tile lateral; y1 = the
elevation of the water in the head control structure above the
center of the tile line at the point in the field where the
equation is being applied and d = the distance from the center
of the tile lateral to the impermeable barrier. Equation (5.7)
is only valid after the water table assumes a level or concave
upward shape and begin to rise uniformly. The time at which
the water table assumes a uniform, concave upward slope and
rises uniformly is the time to. Therefore, the equation does
not apply for the lag-time associated with a subsurface
irrigation system between start-up and attainment of a uniform
water table shape.

Utilizing the data from Figures 5.7 and 5.8 to determine
an appropriate to for Sites C and D, respectively, a calulated
inise of the water table was made. Comparisons of the calulated
éuld the measured rise in the water table at the midpoint

bartween two adjacent tile laterals are given in Figure 5.10 for

105

.m:o_um>eomno .mzcmE L00 mmc__ m__u ucmummvm
cmmzumn >m3n_e .mm_u o_nmu Loam: bouammme new noun—ao_mu o_.m ot:m_u

A8305 mEfi

 

 

 

    
  

 

 
 

 

 

omv OLOV 0mm.) OWN omN OWN OLA“: GLOW 0% O
F L F [P t- b LOHCQOLWSu null O.Q I» O
o32200 Ill Eu 5m.o u
4 nonsmooE Ill =2£E f
l
QT . 1O—
4 so «N u cs 1
l 3; me u or
. E om n 4 tom
0.? 111111
m H .. ...............
W... a TOM.
W O n1 \\ 1
.3; l .
w J 10¢
31 ,
. u\\\ 10m
. .to mgunu 0:.
1 2; non n or 1
nYmL pt an
4 o. ._ w om
l f
l o 35 .2: E0: E0 03 n 803m :8
0.0L 0 BE .2: E0: 60 mm P. oootzm :om105

uogerle elqoi J910M

(we) Jeiueo em LUOJl

106
Sites C and D.

The value for to was the first measured time when the
water table assumed upward movement with a uniform surface
shape during irrigation. The value of to = 30.5 hours for
Site C was where the water table had an approximately level
shape during subsurface irrigation. The corresponding mo
value between laterals was 2.5 cm. The values for the other
parameters on Site C in Equation (5.7) were K =0.51 cm/hr, f :
0.102, d = 45 cm, y1 = 72 cm and L = 10 m (Table 3).

For site D, to = 68 hours when the water table was first
measured having an approximately level and uniform shape during
subsurface irrigation. The corresponding mo value was me = 22
cm. The values for the other parameters in Equation (5.7) are
K = 0.51 cm/hr, f = 0.105, d = 17 cm, y: = 54 cm and L = 20 m
(Table 3).

The values for hydraulic conductivity (K) were taken from
the U.S.D.A., Soil Survey of Tuscola County, Michigan. The
reason these values were used instead of the K values from
Table 1, were that the values represented a more realistic K
value at the actual depths of the tile laterals for Sites C and
D. The K values expressed in Table 1 represent a depth of only
50 cm (20 inches), while the tile laterals were actually
deeper, varing from 66 - 100 cm (26 - 39 inches) depth. Site C
and D contained a Wisner loam having a shallow clay barrier and
1flmerefore the depth of the clay barrier in relation to the tile

Jirteral can have a controlling effect on water movement out of

thss tile.

107

Calculating the height of the water table (m) above the
center of the tile at the midpoint between tile laterals using
Equation (5.7) and comparing the results of the manually
measured water table elevations, it can be seen how close the
measured and calculated values are. In Figure 5.10, Site C
shows very good agreement between the measured and calulated
water table height midway between tile lines until
approximately 250 hours after initial irrigation. Rainfall
then contributed 0.97 cm of water to the soil profile and
caused the measured water table height to increase beyond the
calulated water table height.

Site D shows less agreement between the calculated and
measured values (Figure 5.10). The calculated function shows a
flatter rise in the water table height than the measured over
time. Additional soil moisture from rainfall did not affect
the water table as in Site C. This may have been due to the
deeper depth of the static water table in relation to the soil
surface. The greater depth of the static water table provided
additional storage area and therefore, the affects of rainfall
would not be as great. A second reason for the measured water
table being higher than the calculated may have been the actual
:field hydraulic conductivity (K) may have been greater than

imeasured K used to determine calculated height.

108

Table 3. Parameter values for Figure 5.10, where time
(hours) is referenced from initiation of
irrigation and water table elevation (cm)
referenced from center of tile lateral.

Site C

e ur d Calculated
Time Elevation Time Elevation
(hr) (cm) (hr) (cm) where
30.5 2.5 30.5 2.5 f : 0.102
246.0 40.0 246.0 38.0 L = 10.0 m
385.5 66.0 385.5 52.0 K : 0.51 cm/hr
y1 = 72.0 cm
mo 2 2.5 cm
= 45.0 cm
to = 30.5 hr
t = x value (hr)
m = finding height (cm)
of water table
Site D where
68.0 22.0 68.0 22.0 f = 0.105
282.0 33.0 282.0 26.0 L = 20.0 m
428.0 37.0 428.0 28.0 K : 0.51 cm/hr
y1 = 54.0 cm
mo = 22.0 cm
d = 17.0 cm
to = 68.0 hr
t = x value (hr)
m = finding height (cm)
of water table
Site C Water Stage Recorder Values where
83.0 7.0 83.0 7.0 f = 0.102
96.0 13.5 96.0 9.6 L = 10.0 m

120.0 15.0 120.0 14.2 K = 0.51 cm/hr

144.0 21.5 144.0 18.6 y1 = 72.0 cm

.168.0 24.5 168.0 23.0 mo : 7.0 cm

LL92.0 26.0 192.0 26.6 d : 45.0 cm

216.0 32.0 216.0 30.2 to = 83.0 hr

245.0 37.5 245.0 34.3 t = x value (hr)

264.0 46.5 264.0 37.0 m : finding height (cm)

288.0 52.0 288.0 40.0 of water table

312.0 54.6 312.0 42.3

336.0 58.5 336.0 45.0

360.0 62.5 360.0 47.0

384.0 56.4 384.0 49.0

385.5 56.4 385.5 49.2

109

Measurements of the water table should have been carried
out at more frequent intervals after initial irrigation began.
This would have given additional water table elevation points
for comparison to the calculated elevation height soon after
initiation of irrigation.

Additional observed water table elevations were available
from the continuous record for Site C provided by the water
stage recorders located midway between two tile laterals at the
north edge of the field (Site D had manually measured water
table height data only:).

Data points used from the continuous water stage charts
were taken at about 24 hour intervals. Time was from 83 to
385.5 hours after initiation of irrigation during a fairly
uniform rise in the water table (Table 3). The water stage
recorder values were comparied to the calculated water table
position taken from Equation 5.7.

The results of the comparison between the more frequently
measured and calulated rise in the water table elevation on
Site C show fairly good agreement from 83 to 385.5 hours for
the data obtained from the water stage recorder (Figure 5.11).
At about 250 hours into irrigation the measured and calculated
water table elevations begin to separate. The measured
elevation rose at a more rapid rate due to rainfall of
approximately 0.97 cm, which was not accounted for in the
calculated values.

The results of the measured and calculated water table

elevations are as follows: 1) The measured and calculated

110
lLQJU198

(an)

0')

q-

.u mu_m .mo:_m> boom—so_mo

 

 

  

momcm> mmc__ o__u popuoooc EOLm om_c o_nmu Loumz __.m mesm_d
thDOrC mEE.
om¢ 00¢ 0mm can 0mm CON 0me om— 0% o
L Fll. L. .. . t t .668 .3 ML. 0
32030 .5 3.0 L 2 oo

1 “332200 ..I n 2852 r

1 \ ..S
I o
L Eu 5 I w: \ a.
8; mm n a \
1 E OF H 4 \ Lmncoomc mmBm c303 fiON
\ ummeno nco “3323.8

0 mam 1
1 tom
1 10¢

a
flow

L

1
1 o mtm .2: E0: Eu mm H 8996 zom Om

. a

2: c8253 332:. .0 oxm do new Ltoz ”c0383

 

 

(we) uonoAaga alqm 1310M

111
water table elevations for Site C are in fairly good agreement
as shown in Figure 5.10. The values for Site D are not so
close and reflect discrepancy between the measured and the
calculated rise in the water table height. 2) The results of
earlier and more frequent water table readings from the water
stage recorder on Site C shows good agreement with the
calculated (Figure 5.11). The effect of rainfall is obvious
when the measured water table height increases rapidly as a
result of a 0.97 cm rain; 3) the rise of the water table on
Site D is not as rapid as the rise in Site C. A smaller water
supply was provided at Site D. Site D had a water supply of
about 500 liters per minute (132 gpm) while Site C had 1000
liters per minute (265 gpm) over the approximate same area
being irrigated. The farmer may wish to increase the pumping
rate to increase the rate of the rise of the water table for
Site D and 4) the ability to move water into the soil profile
for the purpose of subsurface irrigation was adequate with the
present tile spacing. The water table continued to rise
despite water loss due to evapotranspiration.

Because of the good agreement between observed and
calculated water table elevations using Equation (5.7), it was
possible to predict the position of the water table in
relation to the length of time irrigation had been on-going
for Sites C and D. Subsurface irrigation is considered a

viable practice with the present tile design and properties.

112
D. Continuous Observations Using Water Stage
Recorders on Site C

Continuous water level recordings using water stage
recorders on Site C cover a time sequence from July 15, when
monitoring was begun, to approximately August 29. July 15,
2:00 p.m., was the beginning of monitoring and this date and
time were referenced as time 0.0. Hourly data was digitized
from the continuous water level trace to form the computer
plots displayed in Figures 5.12 - 5.25 for analysis. A zero
reference level at the top of the head control structure which
housed the valves for controlling the water level was used.
The soil surface, depth of the fluctuating water table, depth
of the clay barrier and depth of the tile laterals were
referenced to this datum. The lateral tile spacing was 10
meters. The figures presented distinguish between the water
table at the north and south end of the site and each display
the water table movement near and/or midway between two
parallel tile laterals. The north end of the field was the
end furthest from the water supply and head control structure.

Rainfall data consisting of an approximate day of
occurrence as recorded by the farmer/manager and cross-checked
at nearby weather stations was plotted. The rainfall amount
(cm) recorded is illustrated with a bar chart when rainfall
occurred in Figures 5.1 through 5.25.

Irrigation was initiated just prior to tasseling of the
corn on July 18, at 8:00 a.m., 66 hours after monitoring of the

water table had begun. The water level in the head control

113
structure was maintained 72 cm above the tile laterals where
monitoring took place. The pumping rate was measured at 1000
liters per minute (265 gal/min) at that time.

At the north end of Site C, monitoring tubes for the
water stage recorders were approximately 70 cm deep in the soil
profile. This depth varied some, depending on the ground
elevation change between the two parallel tile laterals
monitored. The rise in the water table was first sensed 44.5
hours after the initiation of irrigation over the tile lateral.
Midway between the tile lateral, the rise was noted 78 hours
after the initiation of irrigation (Figure 5.12). Reasons for
the lack of immediate response of the water table was due to;
1) the 360 m distance the water had to travel from the water
source and filling of the tile laterals and soil profile around
the tile and 2) the time required to raise the water table
residing below the monitor tubes to a position where
measurements in the monitor tubes could be made.

Figure 5.12 shows the water table above the tile lateral
(dashed line) reaching the monitored depth between laterals of
135 cm, 69 hours after irrigation was begun. In an additional
9 hours, the rise of the water table midway between tile
laterals was registered by the water stage recorder. The 9
hours difference between water table response over as comparied
to midway between laterals represents the difference in time
for the water table to reach the same elevation.

As expected, the water table over the tile lateral always

responded quickest to any change in head at the control

Depth from datum (cm)

114

110- datum = 0.0 Irrigation was initiated July 18. 0 8:00 a.m..
‘ 66 hours after monitoring was begun

 

 

 

 

115*
J
I
120.: 10:30 a.m. diurnal fluctuation
: Soil surface from datum A beginning
. -over tile lateral =- 71.9 cm
125: -between tile lateral - 66.5 cm
1 tile spacing - 10 meters
4
1301
3 69 hours
135 \7[_ initial rise began O 78.0 hrs.
< / after initiation of irrigation:
: irrigation begun /
1 ' /’
140-1 \ x
J J/
11" ____________________
3 ‘~ 44.5 hr
145 4
150‘
4
1
155-4
4
l
160-l
-- over tile lateral
. — between tile lateral

WW
0 20 40 60 80 100 120 140 160 180
Time (hours)
Figure 5.12 Initial water table movement over and

between tile after onset of irrigation
for Site C, north end.

115
structure. The driving force moving water to the middle of
adjacent tile laterals was the hydraulic gradient created when
water came up higher over the tile lateral than midway between
adjacent tile.

The water table continued to rise uniformly for 98 hours
(10:30 am) after the initiation of irrigation and averaged 53
cm (21 inches) below the soil surface at this time (Figure
5.12). Because irrigation was initiated during peak water use
when the corn was beginning to tassel, the downward movement of
the water table (10:30 am in the morning) revealed the
beginning of diurnal fluctuations in late morning due to
evapotranspiration as capillary moisture from the water table
entered the root zone to replace ET.

Figure 5.13 shows the rise of the water table at the
south end of the field, which was quite different than at the
north end (Figure 5.12). The south end was nearest the water
supply and closest to the drainage outlet. The monitoring
tubes for the water stage recorders over and midway between two
parallel tile laterals were 107 cm deep in the soil profile.

When the farmer closed the valve to keep the tile system
from draining, the total drainage system continued to feed
water into this area before irrigation had begun. This
resulted in a gradual rise in the water level both between and
over the tile lines at the lower (south) end of the field.
The water table over the tile lateral (dated line) had a 12.5
cm higher starting-profile than between the tile lines. This

is revealed in figure 5.13, where a gradual rise of the water

Depth from datum (cm)

U1

(11
gr 1 L [L LLLLLLLLLL LLLL LLLI

 

01

()1

116

soil surface from the datum 10:30 on)“
—over tile lateral = 57.0 cm /
~between tile lateral = 53.04 cm //
/
/
2 p.m.,. .I/
/
/
/
/
/
/
/
/ 8:30 a.m.
/
/
/ /
/ I
/
/
/ 9 a.m. 5‘30 pm.
/
/
1 6:30 p.m.
l
J
... 6 .m.
/ P
/ .I
/ /.’
/
l" / ‘/
l, irrigation initially
/ begun July 18, 8 a.m.
/ 66 hrs after monitoring
// was begun
—-—- between tile :ateral
—- ~ over tile lateral

 

 

[IIUIIIITPIIITTIIIIIIIITIIWIII W”IIIIIIIHIIIIIIIIUWIIITIIIWIIIIIIIIIITTIIIIIIIIIIF]

20 4o 60 so 100 120 140 160 180

Time (hours)
Figure 5.13 Initial water table movementover
and. between tile after the initiation
of irrigation on Site C, south end.

117
table was seen before the initiation of irrigation at 10 hours
after the initial monitoring time.

Diurnal fluctuations of the water table began
approximately 98 hours into irrigation, with the water table
being approximately 53 cm (21 inches) below the soil surface
(Figure 5.14). The time on the curves (a.m. and p.m ) refers
to when an inflection point occurred. The water table between
and over tile laterals at the north end of the field rose and
fell in a very rhythmic pattern. This resulted as a
consequence of drawdown attributed to the upward movement of
water from the water table to replace soil moisture lost by
evapotranspiration (ET). The time drawdown began was usually
between 0830 and 1030 hours and fell anywhere from 5 cm to as
much as 15 cm during the day. The rebound of the water table
began between 1730 and 1930 hours. It was interesting to note
that the water profile above and between the tile laterals
responded almost alike in the daily fluctuations. Each day,
the rebound exceeded the rise the previous day by about 2 cm.
The peaks between tile lagged the peaks over tile by about one
hour each day. Rainfall resulted in not much of a rise in the
water table the day it rained (Figure 5.14). Because the water
table rose to a greater height after each decline, a continuous
rise in the water table toward the surface was evident, even
though ET was estimated to be between 0.38 to 0.50 cm/day. The
pumping rate was approximately 1000 liters/minute (265 gal/min)
during this time. This pumping rate was equivalent to an

outflow or ET rate of 0.44 cm/day over the field.

118

0mm

.0 oxm .oco ...to:
..D so oouuoto mo £830. 25
coosson oco ..o>o mcozoaoa: 65:5 .14.... 959d

A9305 oEfi

CNN omN onN oFN om. on— on

bb-h—bn—bhhthbhFh_~thh.Lp—bhb—bbhuhlphpn—hbnth-b-bhthFi—htbpbhbn

I
F

 

(we) ilolulDH
0. C2 0 O.
L 1 1311411 .L‘SZJ L 171 1 1T

0.
.‘T

l

 

of.

Eu m.o "mm
Hanucwmt

mam» cmw3uon Ill
mafia Lo>o iii

 

S OH u mcfiumom onu
Eu m.mm u Hmwuma mama consume
Eu m.~e u antenna can» um>o

TIIIWTIT‘ITT
o in o in O
N N '0 n Vt

ID

in
O

 

O
D

 

[Tlr'rlllllITVITWITIIIIITTTT'IIIIT
O

'0
O)

codum>mao oumusam Hmow

(we) wniap we.” uideg

119

During this same period, the water table fluctuations at
the south end of the field showed a different response (Figure
5.15). Daily fluctuations due to evapotranspiration over the
tile line were almost non- existent, while between the tile
laterals, the water table showed fluctuation, but not as great
as those shown in Figure 5.14 for the other end of the field.

Rainfall contributed to the steady increase of the water
table toward the soil surface without a decrease caused by
evapotranspiration. The effect of rainfall on a rising water
table over and midway between adjacent tile laterals can be
seen in Figures 5.14, 5.16 and 5.17. In Figure 5.14, a 0.5 cm
rainfall 7.7 days (185 hours) into monitoring was enough to
avoid a decrease in the water table due to daily
evapotranspiration. The water level came up more over the tile
than midway between tiles. Figure 5.16, for the north end of
Site C, shows the same result for a 0.97 cm rainfall. The
water table showed a 10 cm upward surge in less than 30
minutes, while midway between the tile laterals, the upward
surge was only 4 cm in 60 minutes. The magnitude of the surge
upward may have resulted from surface or rain water moving down
along the monitor tube during and/or after the rain. The water
surge upward was followed by approximately 8 cm decrease over
the tile lateral and 1 cm decrease midway between tile lines.
The decrease was less than the rise, and the water table
continued to move upward.

The south end of the site for this same time period and

rainfall, Figure 5.17, again showed a similiar response to rain

120

0mm

.0 8m .Em: B 25 £38 .5 .3
nobooto as 233 c302, DESBQJE 2.0 959...

$505 oEc.

 

llDlUlDH
C?
N)

(R
d.

(1113)
C?
If)
JJll—Llililllllijll

C?
to
11

l

L

C?
i\
l

 

O
$41 1

 

E0 D.O n m

.232 0.: .83 III =35...

.E... can

 

.E.ea
\
\
\
\ lllll \
ll \ .E.O
Ea. [\eieecM_ Z
Ego sill
Eon 1. \IL
.690 F— ~
ill] a
\ /\
.E.e a

 

[TUWIT‘TjIUIIT111TT[YFTTIFTVIlrerjijiijTrTj

Ohm omN onN o—N om? on? on—

thbhbbi—bLi-LfibbbbthLbBbbbthLbFl—bthhL—ihbbbbhl-bb—rrbhtLbLPthbbbrbb

.332 0.: c0232.

0
d.

1.0
f’)

O
in

L!)
N

O
N

1.0

O

10
O

O
0

ID
0)

(LUO) wniop uJOJi uideg

121

.oco £6: .0 otm ca 29,32
2: ..o>o oca coozzmn oBB boob; 0:.“ 930E

A9590 oEfi

one on.“V on!a o 3 0mm. Ohm own own 0 PM

 

 

  

 

 

 

 

 

FthhbhuhpLh LPhb.—bbiphbipth—ybi—nbL-htkL—ihlthhbbi-L—bkhipbbbbh—uhnbhhbhb—nthbPhhh OOF

. :8 and u \L \1
L .326. E0 had ..u \ fl
. .4 =2£E \ \1

0 TH \. \\ 1.00
j
”in Rousseau 05 £2 a \ \\ w
O.N1 to Sc» 95:. .E... s 7\ \ 1

1 .6... one. // \\ Tom
a ”+4 . elm-.5 H
D 0.3 \s ...a 92 1
l. .4 \ : 1.

my . : 1mm
D 1 : n
H 061.11 .._ 1
T

\dw/ .11 ...1: 1T0m
w 0...... 1
( 1 T
1 1

. 1 .m.—ND C038 ll WEN.
0 0L mafia Hw>o iii 1
1 liir
1 .\ .. 6.30. 0.: .33 ooatsm zoo 1

0.51. .2... WON.
A .I
. .1llllllllliillllllllllimoumstMelcmwflaisiflmflMm...mofl.

0 mi row

(we) quuap LUOJ; uideg

122

one

.25 £038 .0 Ba

._o._8.o_ 2: $5 £8 00 oocoscg 2.0 83E
A9305 oEfi
one one o 3 own cum can own 0 F.

rthP-th—P-EFB

 

N1

LIJ

(we now
LJJJLJJALJJJJJ

(O

 

h—bippu phthhbb‘pbbbbtbbhbbbbbbebbbnti-trbipbbhbhn—Ephubbb_
Eu end a \L
=3£E Eo had a.
=££E
Aouocfico 05
to Sc» aEsa
Ed 0 _
7 _
Ea A _ /
l \ / —
/ : .E.O\r\
/ \ _ / \
E Po \ , ..\ .
/ a \ / \Ed 0.0.0 ED 0
// E x.» \ /\\ .
/ \ / \\ E6 3
/ \ x1\
(2:... n \\ .6... 2
.ea 2/ 7..\.
/ \/a E o :
/ \\ .Ed n
/l\
.2... 2 e... h 3: 5.23 :2 Bus

 

.832 0.: .33 Ill

Ea hm n coarse zoo

 

I'11T1j1T111—11vr1111111111111111Ijjj

00—

II)
CD

0
O)

1.0
(I)

O
(I)

In
[\

O
l\

L0
(0

(we) wniop Luci; uideq

123
over the tile lateral. The water stage recorder between the
tile laterals jumped its chain on which the float and
counter-weight were hanging and as a result no recordings were
obtained midway between laterals. The water table over the
tile lateral recorded a 5.0 cm upward movement within 30
minutes and decreased by 4.0 cm in almost equal time due to a
0.97 cm rain. It seems apparent that rainfall increases the
water table over the tile lateral, but recedes almost as
quickly as the hydraulic gradient moves the water horizontally
toward the midpoint.

The implications of the magnitude of the rise and fall
implies; 1) the realism of the response additional water can
have on bringing the water table upward in a subsurface
irrigation system; 2) a very quick surge may reflex water
moving along the monitor tubes and 3) the height of the water
table will increase toward the soil surface when irrigation
water is applied at 1000 liter/min (265 gal/min) despite the
diurnal fluctuations downward.

As the water table neared the soil surface, capillary
soil moisture became apparent. When the water table reached
the surface and resulted in a ponded water condition over the
tile lateral, the irrigation pump was shut off (Figure 5.18).

The effect of discontinuing irrigation water and allowing
no drainage, is shown in Figure 5.18. The water table began to
decline. The curves consisted of steeply declining segments
that occurred during the period of evapotranspiration from

approximately 0900 to 1730 hours followed by relatively flat

124

(we) Haiuiog

.96 £6:
.0 0:0 6038329500.ng
B one 203 .6003 0580.600 2.0 930$

A9505 oEfi

 

 

 

 

 

new 0N0 000 00¢ 00¢ 93 nae 091
rLthLleFhrtbtbiplbli—ihhlrirLbE—bbbthFh—tnLiPPlPEbLi—PL hbblhirhbbbl—lhbbb OFF
.332 2: c0053 m n
.832 2: co>o III in
. Ea 00.0 1» =££E I
o T x/ m R:
L T
in
0N 1 no?
11
1 to as...» 9.56 w
1 00050.6 05
. 1 A mine
0 0.1 fl
.1 1
. Woo
Oi n
1 100
1 T
. 1
0 01 n
. low
1 5.1 an T
1 h ./ \ m
0.011 \\ _ T
.1 / \ \ I
O. 51.11111 -..il 1.19.01 3.20.3400 dmditiiflqi ii ii .i ii lll iil iil Lli/flliw. ill iii“
1 .E... : .110K
1 Boga. 22 c0053 coats» :0» H
4| 1. llllllllllllllllllllllllllllllllllllllllllll i...
O.@[ rmm

 

(we) wniop wOJ; uideg

125
segments from approximately 1730 to 0900 hours. A rate of
water table decline of 0.30 cm/hour or 7.2 cm/day was observed.
(A similar rate of decline for the water table later in the
growing season was seen in Figure 5.25, where the water table
fell 0.28 cm per hour or 6.7 cm per day after irrigation was
terminated and no drainage allowed after drainage was
discontinued). Estimated ET averaged 0.40 cm/day (0.16
inches/day) during this particular time. If the drawdown rate
was approximately 7.2 cm/day during average ET rate of 0.40
cm/day, one cm of ET causes the water table to decline
approximately 18 cm in the soil profile. The daily amount of
drawdown depended on the water table elevation, rainfall,
evapotranspiration rate and stage of crop growth. A 0.56 cm
rainfall was recorded during this period and slowed the decline
in the water table. It produced a flattening of the water
table curve followed by a 1.5 cm rise.

At 546 hours into the monitoring process, the farmer
wished to reverse the declining water table. Irrigation was
resumed at 0900 hours on August 7. The water table response is
shown in Figures 5.19 and 5.20.

The response of the water table to irrigation was within
0.5 hr for both ends of the field over the tile lines. The
rise of the water table was not due entirely to the pump
start-up. A rainfall of 1.8 cm occurred which contributed to
the overall rise of the water table. Comparing the rate of
water table rise in Figures 5.19 and 5.20 with the rise in

Figures 5.12 and 5.13 for the beginning of initial irrigation

126

.05 530m .% 0:0
.6032 of of ..o>o :ltoym 0:50
0:0 em as. omcoamoa @300. 08.02, 2.0 930E
A9505 oEfi

0 *0 000 0N0 000 000 0 —0 00¢ 0:1 00¢

 

 

 

 

IhbhLththbLbbLbhb-L—bnbihlplhphLb LFLPL—Lnbbbhhibiblhihbhtnhiphubbbthhhbhhbbi-hbhhnhhthF
1 m Ea 00.0 n m OF —
1 X ..252 .1.
O. T” W Ea 05; fl 1
1 EH0 % =2£E Wmo —
1 /
1 W“ 1
0N.“ 1 / 1
1 x // WOO—
Ud .1 \ / T
m.. D M.” .Edponna .Ed 006 \ .E.u 0nu0/// .Ed 006— .0...
ll \III
M... 1 / .1. \ x -8
D 1 /.._ / \ .Ea. / 1.
...iili O .V.” .56 000; \\ on; .E... 0n“N—/// 1.
\i/ 1 x \ / T.
.3 1 / x // mom
W C Di“ J \.E.a 006.. // fl
. 1 \ .Ed 85 e / é... . 1
0.01.11 x \ Qatoam QESQ .E.a guaf/{0WMnrd o. ”mm
1 0 \ .Ed onfi—J/ 0..
.1 ~\\ / T
T
0 .m.. .5 2K / 100
H 6.132 of .616 ii .1.
0.0L :98»: was camsumn 303.226 no: .953 [0m

wniop LUOJ}
one/\eie eiqoi .ieiom

(we) 0

127

(we) llDJUEDB

 

.220 .6 0cm £16: .0 EB 001.32
2: 83 omcoamoc 208. 0302, 00.0 930.1“.

0 F0 000 own 000

Pth-thPPbb.hbhhrP+-—bbbbrhbhhh~

A9505 0E?

001v 0h... 00¢

hb-tPhFPbbbDPbtPPbLPP-ht—bbbthbnh—bbbppnnbihh

 

 

1 .8... a
0 Tu P
1 .6... on". \
\
1 / \\
OT 7...... . \
.1 / \
1 / \
.v 1 / \l..n.o 8.:
0 1 /\
U .6... 86
0.0.1 .E.o 00am
1 deltas.»
1 aEsd n.
. 11
0 0 1
1
0.x...
0.01

l

V

O

A

‘ 0E0 D." u

0

V
A

O

V
A

O

V
A

O

V
A

O

V
A

l

V
A

9

I
A

v
1 N

V

l.

 

so as 1.. m
=2£E

astound ma: coosuwn ill

HmuwumH 0H0» uo>o iii

0:

00—

00—

IO
0)

O
0!

ID
(I

00

0x

(we) wniop won uideg

128

was therefore not possible. The water table at the south and
north end rose to within 23 and 18 cm from the soil surface,
respectively. Thus the south end showed a higher rise of the
water table due to the rainfall and the hydraulic pressure
created when the water level was raised in the head control
structure.

The water table midway between the tile laterals,
furthest from the control structure of the subsurface
irrigation system (Figure 5.20), took 12 hours to respond to
resumed pumping when the water table was initially 37.5 cm
below the average soil surface. Forty hours after irrigation
was resumed, the water table between the tile had risen a total
of 8.0 cm, which was at a rate of 0.2 cm/hr (4.8 cm/day). The
water table between tile lateral never peaked as high as that
over the tile laterals. The maximum height wver the tile was
18 cm below the soil surface as compared to 27 cm between tile.
A 9 cm difference (3.5 inches) in height between the two water
table curves existed. A second event involved a second
pump start-up event later in the growing season, 35.7 days
(856.5 hours) into the monitoring of Site C, is shown in
Figures 5.21 and 5.22. An almost immediate response by the
water table over the tile lines could be observed at both ends
of the field. Midway between the adjacent laterals, the rate
of rise was slower. An actual decline or receding was observed
until 5 hours into irrigation, at which point the water table

began to rise.

129

.08 £30m .0 otm .poEemoL
cozamtt can; £830. 2: :83qu
3 0800800 mo ..o>o 957.03 NW0 @5911

$305 oEP

 

 

 

 

0N0 000 000 000 owm 0N0 000
FFLbrbLL—bhtt bybhiniP1—1P blbinPl-bhhhi—ibbibiFihlPL—leLFCLLPLPhbhtL—bL-ph—
.V. ON—
.h E0 ”ZN " W\\ .5 Oak
=2£E ~\\k 711.! n F P
M. \/--// xiii; o:
O.N \& \ //.--////Il
U0 M“ 56013:» \ \ melts» l.
D 0.0 .152. .. \ 95a 00 0
w. .6... 2 \
my \3 \\ 00
NH O .5... 8p—\\0 //\\
\l .6... 2.4.x ..5. 88. cm
W . _
(0 Ow mm
of , _ cm
H“ ¥E.e.
. 1% 0m.
3. 5 a.
“0030. 05 ..o>o il ON.
1‘

“E1

“0032 20 coosuon i.l

(we) wniop LUOJJ, uideg

130

.08

to: .0 arm .moEsmoc
.550 :1; can; £01.32 25
coozzon B vocaano ma ..o>o 0251001.

NN.m messes

@305 9:0.

0N0 000 000

—tPL_th~be—

0v0 0N0 000

 

 

1

OT

OT.

.1

.1

moo?

nii. .1

m1 1
D

”041.1

1

1m, 1

i

w Om1

/l\ .1

1

.1

of

O?

1

IA

E0 MWN “
=Owcmnv1—

c300
“ace aEaa

\ \

 

\
a
_
.
_
_
_
_

_

_

_

I —

/ _

 

0.01.11

 

 

nnpHL_1—_iPtnbththLt—~b—nbbptP-b_thh—_hkh

m8: 8&0
///1////.
./1//
. . . VIA?
E o 00.0 .I/..//
//
// ./
I.
fill].
.00: 0.0mm «0

voEauE unmaEsa

 

.232 2: c0233
.332 2: ..o>o Ill

 

l
l
i
l
l
l
l
l
l
l
l
l
l
l
l
l

 

[0" I..."

.1932 20 ..o>o weatmu =0» .

l
l

l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
l
I1171IIITTIITTTTFTTTIITUfl17‘11rIIIIIVVWVjTI ITITTTWW

.332 2: .6253 cactus =0»

 

09

IO
F
F

O

V'-

IO
0

0
0

F

In
03

O
G

If)
0

O
0

ID
I\

O
IN

I")
(D

F

(we) wniop LUOJ} uzdeq

131

.vco renew .0 ou_m .o__u ozu bo>o
omm:_mcv 0cm __m0c_mc ou omcodmoc o—nmu Loam: m~.m obem_u

A9505 oEP

men 0mm. 00m. 000 000 0¢0 0N0 000 000

 

 

 

 

 

nthbbbe—PbbhbnthDPbbhhbhb—PEPHbPhP bbbbbbeP—P PDBL-phi-bbbmbbbh-bbbhpbbhb m0"
1 W.“ 1
1 “M" w
J .
O. _..1 .832 20 .06 III "H“ 11.100—
1 "N“ w
1 Eeofls v1 I. W
1 .Ed 006 7 "Nu \ ll: ~_ 1. 00
OT 111) 1 , .«1 \ x .1.
1 / 0 / Wm 1
H0 1 so on": t x x no“ 1
D 0.01 a... one / ..x 1». n
11. 1 x 0 .lu .imxw
u x _ "N“ 1
I1 1
11H 0 7.11 r 1f/ ~ "Nu =3£E How
o o 05‘ . 1
w 0.01 .5... on”: / .6... 4”» LE... and v“; WON.
( 1 x _ w
. 1 once 58 x _ 1
O 0.1 u _ . u // __ h
.1 I
1 . x _ n00
E... on.» 11.
. - .5... on.» .1.
O 5.1 0E... D 0826 c0000.:— 1
1 0:0 pecoao 50.5 W00
1 1
11i.i.ii.i.iumNWMMfiflmMi. lllllllllllllllllll i1n
o m1 _ 1.11m

ago/\eie eiqozi 1910M

wniop wen

(we) 0

132

The effect of a rainfall event in Figures 5.23 and 5.24
shows a rapid rise in the water table at both ends of the field
when 5.0 cm were added. Irrigation was ongoing during this
time. The water table rose 42.5 cm over the tile lateral in 95
hours. The water table came within 5 cm of the surface at the
south end of the field (Figure 5.23) and actually ponded water
over and between the tile laterals at the north end of the
field (Figure 5.24). The water table between and over the tile
lateral rose uniformly. At this point in time, the farmer
ceased irrigation for the second time and opened the tile main
to drain water from the soil profile. The amount of drainage
water was not recorded. Skaggs (1981) discussed this situation
when he noted that rainfall can be lost because of the lack of
storage available in the soil profile as the water table nears
the surface. Water can leave as excess run-off or be lost as
overflow of the water control structures.

Once irrigation had ceased and the drain was opened to
remove excess rainfall, within a 24 hour period the water
table at the south end of the field over the tile lateral
(Figure 5.23) fell 17 cm. The rate was 0.7 cm/hr or twice the
rate at the furthest end of the field (0.3 cm/hr) on the same
tile lateral. Once the drainage valve was closed, the water
table first rebounded, then continued to recede. The water
table receded at a rate of 19 cm in 19 hours which was twice
the rate midway between tile.

During this same period of time the water table at the

north end of the field decreased 7.2 cm both over and between

133

.ucm coco: .u muwm .mmmcwegu
use __mochg op mmcoammg apnea swam: ¢~.m mnemwu

$505 08:.

n¢h nun won new mac nfiw

Lpbhbbbbbb—lh-bnbhtbbebl-hDbl-b—bblbhbbhbb

bPthbhhh—

on» now

bB-htbl—b

 

0.
Ln

 

.830. 0.: c0235 III
.232 2: ..26 I I

coco—o 52v

 

/

'i II}

.832 2: ..o>o oootsm =0»

.5... w

 

.932 2: 5233 38.5» =0»

coco
Soup

 

 

'01010101014

01:1010

‘
A

,3

\
Cr,

1010101

I0

E0 0.“ '

101010

..3501.

0

V
A

0

01

V
A

02010

0

"
AA

0

5'
a
0101

~
I v V
A

I?
a.
}

 

Illllilll

non

oc—

\\\.//y 00?

mm

on

on

on

wniop mm;
(we) uogerge elqoi 1930M

134
laterals (0.3 cm/hr) in response to the drain being open
(Figure 5.24). The water table continued to decreased when the
drainage value was closed after 24 hours due to redistribution
of water midway between tile lateral toward the tile lines and
evapotranspiration and other possible losses. The water table
decreased an additional 14 cm in 28 hours (0.5 cm/hr) before it
began to flatten.

The downward diurnal water table drop under closed drain
conditions at the north end were approximately 4 to 5 cm per
day over and between the tile laterals (Figure 5.25). No
further water input occurred after 790 hours (August 17). The

water table was allowed to decrease through ET and with no

drainage.

a 5 4. For subsurface
irrigation with a constant evapotranspiration rate at the
surface, the steady state water table takes the general shape
given in Figure 5.9. An algebaic expression previously
mentioned, Equation 5.5 (page 100), which describe the water
table can be used to determine the tile spacing required to
give a maximum difference, hl - hz, in the water table
elevation directly over (hl) and midway between (hz) the tile
laterals. hl and hz equal the distances of the water table
above the impermeable layer over the tile and midway between
the tile lines, respectively. This equation can then be used
to assess; 1) the desired tile spacing (L) for drainage if the
hydraulic conductivity (K), the evapotranspiration rate (u) and

the expected water table height above the impervious barrier

135

.ecm cute: .0 as?»

.Hm new: umum_oommm czouzmgu open» new»:

m~.m wgsmwm

 

A9505 Etc.

000 onm own can Ohh omh onh can can Ohm can can
thhhnbleb—hbhyhbbbnb—nbthbthbbebbD—nbDh-hhébhhb-nbhhbbbbbbnthh.bhbbbtr-b-beh—bb Dth-P—bbbn-hzb “F"
.232 2: 5232. II. n

I n
.232 0.: 1.26 .110:
h
2.3.0 58“. mano—
1=o n
COWOUEm FOOw
coco m
£96 1 ma
om
no
on
.33» DB

 

 

.830. 2: 33 coat...» =0»

‘IQ‘Iillllllllllll!llllllllil

.832 05 :33qu coats» =0»

 

 

   

 

cs.

 

ITIUTTrIjTIT ‘T'fii‘l‘ITVITII‘TTTIT'

O
‘D

O
b

If)
CD

wmop mm;
(we) uoperle elqm 1910M

136
either over or between tile laterals is known; 2) the
expected water table response over and midway between the tile
laterals of a drainage system already installed to determine if
the present spacing is adequate and 3) the tile lateral
spacing for subsurface irrigation when the water table
elevation difference over and midway between the tile laterals
is set based on allowable variation for a particular crop.

The design spacing for a particular soil can be computed
by choosing an optimum water table depth and a maximum
allowable variation (Skaggs et.al., 1972). Williamson and Kris
(1970) give yield response of agricultural crops to a variety
of constant water table depths. They showed optimum water
table depths of 76 and 90 cm from the surface for corn on a
loam and silt loam, respectively.

Data from Figure 5.14 was used in Equation 5.5 to
evaluate the water table position with the present design
and/or layout of Site C. An average K value of 4.2 cm/hr
(Table 1) was determined from a velocity head permeameter. An
average evapotranspiration rate of 0.5 cm/day was determined
using a modified Penmann equation (Vitosh, 1984). The hi and
h2 values were determined from Figure 5.14 (hi = 76.7 cm, h2 =
71.7). A theoretical tile spacing of 7.7 meters was calulated
where an actual spacing of 10 meters existed in the field.

This agrees fairly well with the tile spacing that was
presently in the field. The 10 m spacing showed that the
water table could achieve a continuous rise dispite the loss of

water through evapotranspiration. This was seen in Figure

137
5.14, where the fluctuations during high ET demand decreased
the water table height in the soil profile, but the continuance
of irrigation through the low ET time of day, resulted in the

water table increasing approximately 2 cm more than the diurnal

decrease from the previous day.

138
E. Yield Results

Yield data for the irrigated and non-irrigated check plots
were recorded in Table 4 for each field site. Table 5 shows
the plant population at emergence for the irrigated and
non-irrigated plots. Table 6 and 7 represent the crop type,
variety, fertilizer and herbicide program, the soil ph, and
available K20 and P205, respectively, for each site.

Site A had an irrigated and non-irrigated check plot for
comparison (Figure 4.2). Plant population at emergence was
66,700 plants/ha, and the fertilizer and herbicide program was
the same. The irrigated plot had a yield of 12,675 Kg/ha
versus the non-irrigated plot yield of 10,955 Kg/ha (measured 0
15.5% moisture). The difference between the two plots, 1920
Kg/ha, was a 17.5% higher yield for the irrigated versus the
non-irrigated plot. This was a significant increase in yield
resulting from the addition of water by subsurface irrigation.

The sugarbeet yield data for Site B was recorded in Table
4. Plant population at emergence was 53,000 plants/ha for both
the irrigated and non-irrigated plots (Table 5). There were
four plots harvested, with one being a non-irrigated plot
(Figure 4.5). Plot one represents the zone over and near the
tile lateral, including the area of puddled water at the north
end of the field and capillary soil moisture visible at the
surface at the south end of the field. Four rows of sugarbeets

(76 cm row spacing) on Table 4.

139

Table 4. Yield data (kg/ha 0 15.5% moisture, corn only).

 

Field sites Irrigated Non-irrigated
A 12,875 10,955
B (plot 1) 63,255 * -
(plot 2) 64,934 * -
(plot 3) 57,545 * -
(plot 4) “ 52,171 *
C (plot 1) 10,092 -
(Plot 2) 11,590 -
(plot 3) - 10,574
D 11,945 -

* sugarbeets.

Table 5. Plant population at emergence

 

 

(Plants/ha).
Field sites Irrigated Non-irrigated
A 66,700 66,700
B 53,000 53,000
C (Plot 1) 66,900 -
(plot 2) 63,400 —
(Plot 3) - 63,400

D 65,200 -

Table 6.

Crop type

Variety

Fertilizer Program

Nitrogen(actual)
P2 05
K20

Herbicide Program

Atrazine
Blade X
Sutan
Formula 40
Brandville
Anthazine

140

Crop type, variety, fertilizer
and herbicide program.

Field Sites

A B C D
corn sugarbeets corn corn
Pioneer -* Stauffer Stauffer
3744 B606 5650
(kg/ha)
293 364 155 155
49 171 124 124
141 364 234 234
(liters/ha)
1.2 -* - -
3.5 - - -
4.7 - - -
- - 0.6 0 3
- - 0.6 0 6
.. _. 4.7 ..

* Not available from the farmer.

Table 7. Soil ph and available soil nutrients of K20

and P205.
”£155 ----- £311 """"" EQIIQQI; """"" E51118; """""
“Eff? ________ ‘3? _________ 89-355932 _____ E’ES’iJEEZE‘EZHH
A 6.5 550 690
B 7.2 490 206
C 6.1 426 6.0
D 6.0 122 75

Note: Nitrogen not available from soil lab.

141
each side of the lateral were harvested over two parallel
laterals, for a total of 16 rows of beets in plot one.

Plot two represents the zone three to nine meters from
two parallel tile laterals. The zone consisted of 16 rows.
Eight rows were harvested along each lateral as shown in
Figure 4.5. No surface moisture was observed in plot two
which contained the area five meters off the tile laterals
where no water table was found. Plot three was the 16 rows
midway between the two laterals beginning nine meters from
each of the two parallel tile laterals, or the area not
included in plots one and two. At no time was a water table
observed within 77 cm (bottom of monitoring tubes) from the
soil surface in plot three. The non-irrigated plot was plot
four.

Table 4 shows that the sugarbeet yield in plot one,
closest to the tile lateral, was 2.6% less than the yield in
plot two. This difference was not considered significant. It
was possible, that the sugarbeets in at least some of the rows
in plot one were too wet, which adversely affected their yield.
Loomis and Haddock (1967) and Archibald and Haddock (1959)
concluded that sugarbeets growing in the presence of lower
available nitrogen and excessive soil moisture will have
depressed yields. The excessive soil moisture which deve10ped
over the tile laterals could have caused a loss of soil
nitrogen through denitrification. Observations in the color of
the sugarbeet canopy during the growing season revealed a

yellowing of the plants over the tile laterals compared with

142
plants midway between the laterals. This yellowing effect was
likely a result of a lower available soil nitrogen or a lower
uptake of nitrogen by the plant.

Plot two had a larger yield than plot three by
approximately 11.4%. It was surmised that the subsurface
water table existed beneath a portion of plot two but dropped
below the monitoring tubes five meters from the tile lateral.
The water monitoring tubes placed five and ten meters out from
the two parallel tile laterals never showed a water table. The
yield advantage of plot two over plots one and three may have
resulted from available soil moisture, available soil nitrogen
being present and not removed by denitrification, and proper
soil aeration.

Plot four had the lowest yield at 52,171 kg/ha, 16.6%
less than plot two, the highest yielding plot. Plot three had
no known water table and thus was expected to be like a
non-irrigated area. However, plot three still exhibited a
higher yield than plot four by approximately 5,374 kg/ha,
9.3%.

The fertilizer program was similar for the whole field
(Table 6) and the fertility samples were not taken separately
for the two soils but were a composite of the whole field.
Table 5 shows the soil ph and available K20 and P205 for Site
B.

Site C had two irrigated and one non-irrigated plot
(Figure 4.6). Plots 1 and 2 had plant populations at

emergence of 66,900 and 63,400 plants/ha, respectively, and

143
the non-irrigated plot (plot 3) had a plant population at
emergence of 63,400. The purpose for the differences in plant
populations was to look at the affect subsurface irrigation
would make on yield.

The results of the corn yield from two irrigated and one
non-irrigated plot for Site C are recorded in Table 4. Plot 1,
irrigated and having a higher plant population than plot 2,
yielded less than plot 2 by approximately 1500 kg/ha. The
non-irrigated plot, having an equal plant population as plot 2,
yielded less than plot 2 by approximately 1000 kg/ha, but out
yielded plot 1, which was irrigated and had a higher plant
population, by approximately 462 kg/ha. Slightly increased
plant populations greater than 63,400 plants/ha resulted in no
increase in corn yield. The yield difference between the
irrigated and non-irrigated plots was not that great.

Site D had no non-irrigated plot. The yield for the
irrigated check was 11,945 kg/ha obtained with a plant

population at emergence of 65,200 plants/ha.

VI. Conclusions

The water table response during subsurface irrigation and
drainage was investigated successfully on three of four sites.
Water table response was affected by horizontal hydraulic
conductivity, irrigation input rates, tile spacing,
evapotranspiration (ET) and soil texture.

Horizontal hydraulic conductivity was a major component in
water movement when attempting to establish a uniform water table
over and between adjacent tile laterals. It was shown that a
saturated horizontal hydraulic conductivity rate 0.3 cm per hour
did not provide sufficent flow to establish a water table in a
clay loam soil. Subsurface irrigation would not be recommended
for similar hydraulic conductivity rates. Soils with horizontal
hydraulic conductivities of approximately 4.2 cm per hour work
well for subsurface irrigation with tile spacings at 10 meters.

Hydraulic conductivity rates for similar soil textures were
shown to be different. The higher rates allow for wider tile
spacing. Because of different hydraulic conductivities between
field sites, an independent evaluation should be conducted for
each field when designing for subsurface irrigation. Narrower or
wider tile spacing will result, depending of the ability of the
soil profile to conduct water horizontally when establishing a
water table under the root zone of a growing crop with subsurface
irrigation.

A reliable method of monitoring the water table is

essential for knowing the depth from the water table to the root

144

145
zone, and whether the water table is rising, falling or at
equilibrium.

Monitor tubes should be placed at the tile lateral, and
midway between tile laterals because this area was shown to
respond the slowest. The tubes should be placed to a depth below
the depth of the tile lateral, wrapped with a filter material and
capped on the bottom to keep out sand and fine silt.

Drainage tile spaced at 10 meters in Wisner loam lowered the
water table seven cm between tile lateral in 24 hours when the
drain outlet was opened. For subsurface irrigation conditions in
which the water table is held at shallow depth, the application
of rainfall will result in a significant rise in the water table.
Thus, subsurface irrigation systems will have to be designed such
that adequate drainage is provided. Therefore, narrower tile
spacing, larger tile mains and larger drainage coefficient need
to be considered for removal of large amounts of water quickly.
This decreases the risk of flooding and crop damage.

Presumably due to ET, diurnal fluctuations of the water
table were observed as a sinuous rythmic falling and rising of
the water table. This sinuousity makes it essential that manual
water table observations always be made at the same time each
day. Then observed trends will reveal a truer vertical water
table movement on the sinuous curve.

An increase in the water table between adjacent tile
lateral over the daily diurnal fluctuations caused by
evapotranspiration is a positive indicator during subsurface

irrigation. The irrigation input rate and tile spacing are

146
adequate to increase the water table response toward the root
zone to irrigate the crop.

Rainfall events brought the water table to the soil surface
during subsurface irrigation. A water table nearing the soil
surface decreases storage for infiltrating rain water. One
management approach would be to allow the water table to
fluctuate over a range to provide for greater utilization of
rainfall. Allowing a decrease of the water table through ET to a
predetermined depth from the soil surface results in greater
storage of rain water. The water table must not be allowed to
decrease to a depth where it can not be raised again. This
practice provides a management tool for decreasing the likelihood
of damage to the crop due to flooding. It will result in
decrease irrigation water pumped when rain water is stored in the
soil profile for the crop.

Good agreement was found between the calulated and actual
field tile spacing for a given water table variation from over
the lateral to half way between laterals using the approximate
steady state approach. This approach, therefore, provides a
means of designing tile spacing when a maximum difference in
water table height above the clay barrier at a position over and
midway between the tile laterals is known for a specific crop.

The yield response of corn and sugarbeets to subsurface
irrigation was investigated. Results revealed that a granular
soil texture produced a greater yield increase due to irrigation
than heavier clay soils. Corn yields increased 17.5% over

non-irrigated yield on a granular soil texture. Irrigated corn

147
yield showed a 13.0% decrease when corn population was increased
5% on a sandy clay loam. Sugarbeets on clay loam soil showed a
greater difference (18.6%) between irrigated and non-irrigated
yield than corn (6.6%). It would be more beneficial to
establish sugarbeets on the heavier clay soils than corn under
subsurface irrigation. Benefits of subsurface irrigation need to
be evaluated against installation, operation and maintence cost
and the financial return on crop yields in the established crop

rotation.

VII. Recommendations

1. Establish a method and key locations which measure the
average water table position in relation to the root zone of a
particular crop and is easy to use by the irrigator.

2. Design a subsurface irrigation - drainage system based on
horizontal hydraulic conductivity rates, water usage of crop
grown, uniformity of the water table over as compared to midway
between tile lateral and timely removable of excess water
through drainage.

3. Timing of subsurface irrigation be adopted for each site
based on crop need, rate of saturated horizontal water
movement, minimum and maximum depth of the water table from
the soil surface and length of time needed to build a water
table.

4. Saturated horizontal hydraulic conductivity measurements
be conducted on each soil horizon to the depth of the clay
barrier and thus provide understanding of the region where
saturated horizontal water flow occurs.

5. Monitor tubes be installed at close intervals in the
region between and over tile laterals to provide a knowledge
of the water table movement caused by rainfall,
evapotranspiration, irrigation and drainage of excess water.
6. Drainmod be run on sites to characterize subsurface
irrigation flux and the response of the soil water regime to
various combinations of surface and subsurface water

management.

148

LIST OF REFERENCES

LIST OF REFERENCES CITED

Allison, F.E. 1965. Evaluation of incoming and
outgoing processes that affect soil nitrogen. Soil
Nitrogen Agronomy Monograph No. 10. Amer. Soc. of
Agron. pp. 578-606.

Allmaras, R.R., A.L. Black and R.W. Rickman. 1973.
Tillage, Soil environment and root growth.
Proceedings of the National Conservation Tillage
Conference, Des Moines, Iowa, pp 62-66.

Archibald, D.B. and J.L. Haddock. 1959. Irrigation
practices as it affects fertilizer requirement,
quality and yield of sugarbeets. J. Amer. Soc.
Sugarbeet Technol. 10:344-355.

Benz, L.C., E.J. Doering, G.R. Reichman. 1981.
Water table management saves water and energy.
Trans. Amer. Soc. Agri. Eng., Vol. 24(2):995-1001.

Bloodworth, M.E., C.A. Burleson and W.R. Cowley.
1956. Root distribution of some irrigation crops
using undisrupted soil cores. Agri. J. 50:317-320.

Bouwer, H. 1959. Theoretical aspects of flow above the
water table in tile drainage of shallow homogeneous
soils. Soil Sci. Soc. Am. Proc. 23:260-263.

Bouwer, H. and van Schilfgraade, J. 1963.
Simplified method of predicting fall of water table in
drained lands. Trans. Amer. Soc. Agri. Eng., Vol.
6:(4)268-291, 296.

Bouwer, H. 1966. Developing design requirements for
parallel drains. Proceedings of the Drainage for
Efficient Crop Production Conference, Amer. Soc.
Agri. Eng., St. Joseph, MI 49085.

Bravo, N.M. and 6.0. Schwab. 1975. Effect of openings
on inflow into corrugated drains. ASAE Paper No.
75-2525, Amer. Soc. Agri. Eng., St. Joseph, MI 49065.

Brewbaker, H.E. 1934. Studies of irrigation
methods for sugarbeets in northern Colorado. J.
Amer. Soc. Agro. 26:222-231.

Chandra, K. 1973. Dynamics of corn root growth as
affected by compact subsoil and its influence on crop
response to irrigation. Unpublished Ph.D., Das,
University of California, Davis.

-149

150

Childs, E.C. 1947. The water table, equipotentials,
and streamlines in drain lands. v. The Moving
Water Table. Soil Sci. 63:361-376.

Criddle, W.E., and C. Kalisvaart. 1967.
Subirrigation Systems in Irrigation of Agricultural
Lands, Robert M. Hagan, Howard R. Haise, and Talcott
W. Edministed<ed.), Agronomy 11:905-921. Amer.
Soc. Agron., Madison, WI.

Doering, E.J. 1965. Change of permeability with time.
Proceedings of the Drainage for Efficient Crop
Production Conference, ASAE, St. Joseph, MI 49085.

Erickson, A.E. 1965. Short-term oxygen deficiencies
and plant response. Proceedings of the Drainage for
Efficient Crop Production Conference. ASAE, St. Joseph,
MI. 49085.

Evans R., R.W. Skaggs, and W.N. Reynolds. 1985.
Agricultural water management for coastal plain soils.
North Carolina Agricultural Extension Service, North
Carolina State University, Raleigh, North Carolina.

Finkel, H.J. 1982. Pipe Flow. Handbook of Irrigation
Technology CRC Press, Inc., Vol.(11:172-166.

 

Follett, R.F., G.A. Reichman, E.J. Doering, and L.C. Benz.
1974. Effect of irrigation and water-table depth
on crop yields. Agron. J., 66:304-308.

Fox, R.L., J.T. Phelan and W.D. Criddle. 1956. Design
of subsurface irrigation systems. Agr. Eng.
37(2):103-108.

Grable, A.R. 1966. Soil aeration and plant growth.
Adv. Agron. 18:57-106.

Gilley, J.R. and 0.6. Watts. 1977. Possible energy
saving in irrigation. Journal of the Irrigation and
Drainage Division of ASCE, 103(Ir4):445-455.

Haddock, J.L. 1947. The influence of plant population,
soil moisture, and nitrogen fertilization on the sugar
content and yield of sugarbeets. Agron. J. 41:79-84.

Hardjoamidjojo, S. and R.W. Skaggs. 1982. Predicting
the effects of drainage systems on corn yields.
Agricultural Water Management. Vol. 5:127-144.

151

Harris, C.I., H.T. Erickson, N.K. Ellis, J.E. Larson.
1962. Water level control in organic soil and related
to subsidence rate, crop yield and response to
nitrogen. Soil Science Vol. 94:158-161.

Harris, D.G. and C.H.M. van Bavel. 1957a. Growth yield
and water adsorption of tobacco plants as affected
by the composition of the root atmosphere. Agron. J.
49:11-14,l76-184.

Harris, D.G., and C.H.M. van Bavel. 1957b. Root
respiration of tobacco, corn, and cotton plants.
Agron. J. 49:182-184.

Henderson, D.W., E.J. Hills, R.S. Loomis, and E.F. Nourse.
1968. Soil moisture conditions, nutrient uptake, and
growth of sugarbeets as related to method of
irrigation of an organic soil. J. Amer. Soc. Sugarbeet
Technol. Vol. 15:35-47.

Hiler, E.A. 1969. Quantitative evaluation of crop-
drainage requirements. Trans. Amer. Soc. Agr. Eng.,
Vol. 12(3):499-505.

Hiler, E.A., R.N. Clark, and L.J. Glass. 1971. Effect
of water table height on soil aeration and crop
response. Trans. Amer. Soc. Agr. Eng.,

Vol. 14(5):679-882.

Hillel, D. 1980. Fundamentals of Soil Physics, Dept.
of Plant and Soil Science. Univ. of Mass., Amherst,
Mass. Academic Press, pp. 147-150.

Hillel, D. 1982. Introduction to Soil Physics, Dept.
of Plant and Soil Sciénce. Univ. of Mass., Amherst,
Mass. Academic Press.

Hooghoudt, 8.3. 1937. Bifdregon tot de kennis van
eenige natuurkundige grootheden van degrond, 6. Versl.
Landb. 0nd. 43:461-676.

Jensen, M.E. 1980. Design and operation of farm
irrigation systems. ASAE. Monograph, No. 3. Chapter
7, pp 242. Drainage Requirements and Systems.

 

Johnston, W.R., F. Ittihidieh, R.M. Daum, A.R. Pillsbury.
1965. Nitrogen and phosphorus in tile drainage
effluent. Soil Sci. Soc. Am. Proc. 29:287-289.

Kirkham, D. 1958. Seepage of steady rainfall through
soil into drains. American Geophysical Union
Transactions, 39:892-908.

152

Kramer, P.J. 1965. Effects of deficient aeration on
the roots of plants. Proceedings of the Drainage for
Efficient Crop Production Conference. ASAE, St. Joseph,
MI. 49085.

Kriz, G.J. and R.W. Skaggs. 1973. Water management
using subsurface drains. Journal of Soil and Water
Conservation. Sept.-Oct., pp. 216-218.

Larson, D.L. and D.D. Fangmeier. 1978. Energy in
irrigation crop production. Trans. Amer. Soc. Agri.
Eng., Vol. 21(6):1075-1080.

Lemon, E.R. and Erickson, A.E. 1952. The measurement of
oxygen diffusion in the soil with a platinum
microelectrod. Soil Sci. Soc. Am. Proc. 16:160-163.

Lemon, E.R. and Wiegand, C.L. 1962. Soil aeration and
plant root relations: II. Root respiration. Agron.
J. 54:171-175.

Letey, J. and L.H. Stolzy. 1967. Limiting distance
between root and gas phase for adequate oxygen supply.
Soil Sci. 103:404-409.

Linville, D.E. August, 1983. Impact evaluation on
increased water use by Agriculture in Michigan.
Research Report no. 449, Michigan State University,
East Lansing, MI, 48824.

Long, W.F. 1965. Drainage design for irrigated lands.
Proceedings of the Drainage for Efficient Crop
Production Conference, ASAE, St. Joseph, MI. 49085.

Loomis, R.S. and J.L. Haddock. 1967. Sugar, oil, and
fiber crops. Part 1- Sugarbeets in Irrigation of
Agriculture Lands, Robert M. Hagan, Howard R. Haise,
and Talcott W. Edminister(ed.), Agronomy 11:640-648.
Amer. Soc. Agron., Madison, WI.

Lucas, R. and M. Vitosh. Dec. 1978. High corn
yields with irrigation. Farm Service Series,
Cooperative Extension Service, Michigan State
University.

Luthin, J.N. 1959. The falling water table in tile
drainage. II. Proposed criteria for spacing tile
drains. Trans. Amer. Soc. Agri. Eng.,

Vol. 2:(l)44-47,51.

153

Massey, R.C., R.W. Skaggs and R.E. Sneed. 1983. Energy
requirements of irrigation systems- Subirrigation
versus sprinkler irrigation. Trans. Amer. Soc. Agri.
Eng., Vol. 26:126-133.

Mengel, D.B. and S.A. Barber. 1974. Development and
distribution of the corn root system under field
conditions. Agronomy Journal 66:341-344.

Michigan Data. 1982. National Resources Inventory.
USDA, Soil Conservation Service, 1405 S. Harrison
Road, East Lansing, MI 46823.

Moody, W.T. 1966. Nonlinear differential equation of
drain spacing. Journal of the Irrigation and Drainage
Division, ASCE, Vol. 92, NOIRZ, Proc. Paper 4835, pp.
1-9, June.

Neal, J.H. 1934. Spacing and depth of tile drains.
Agricultural Engineering. Vol. 15, pp. 194-197,
229-232.

Passioura, J.B. 1981. Water collection by roots,
photosynthesis and biochemistry of drought
resistance in plants. Academic Press Australia.

Patrick, W.H. and R. Wyatt. 1964. Soil nitrogen loss
as a result of alternate submergence and drying.
Soil Sci. Soc. Am. Proc. 28:647-653.

Purvis, A.C. and R.E. Williamson. 1972. Effects of
flooding and gaseous composition of the root
environment on growth of corn. Agron. J.
64(5):674-678.

Reeve, R.C. 1982. Soil permeability--Auger-hole method,
Technical Note No. 2.118, July, Advanced Drainage
System, Inc., Columbus, Ohio 43221.

Reichman, G.A., E.J. Doering, L.C. Benz, and R.E. Follett.
1977. Effects of water-table depth and irrigation on
sugarbeet yield and quality. J. Amer. Soc. Sugarbeet
Technol. 19:275-287.

Reicosky, D.C., R.J. Millington, A. Klute, and D.B.
Peters. 1972. Patterns of water uptake and root
distribution of soybeans (Glycin max) in the presence
of a water table. Agron. J. 64:292-297.

Renfro, G.M., Jr. 1955. Applying water under the
surface of the ground. U.S.D.A. Yearbook.

154

Richards, L.A. 1931. Capillary conductivity of liquid
through porous media, Physics New York, 1:318-333.

Robins, J.S., C.E. Nelson, and C.E. Domingo. 1956.
Some effects of excess water application on
utilization of applied nitrogen by sugarbeets.
J. Amer. Soc. Sugarbeet Technol. 9:180-188.

Schwab, 6.0., Frevert, R.K., Edminster, T.W. and Barnes,
K.K. 1981. Soil and Water Conservation
Engineering, Thrid EdItion, John Wiley and Sons, Inc.

 

 

Sieben, W.H. 1964. Het verband tussen ontwatering en
opbrengst bij de jonge zavelgronden in de
Noordoostpolder. Van Zee tot Land. 40, Tjeenk Willink
V, Zwolle, The Netherlands.

Skaggs, R.W. 1978. A water management model for
shallow water table soils. Report No. 134, Water
Resources Research Institute, N.C. State Unversity,
Raleigh, pp 17.

Skaggs, R.W. 1979. A water management model for
shallow water table soils. Report No. 134, Water
Resource Research Institute, N.C. State University,
Raleigh, pp. 178.

Skaggs, R.W. 1981. Water movement factors important to
the design and operation of subirrigation systems.
Trans. Amer. Soc. Agri. Eng., Vol. 24(6):1553-1561.

Skaggs, R.W., G.J. Kriz and R. Bernal. 1972. Irrigation
through subsurface drains. Journal of the Irrigation
and Drainage Division, ASCE, 90:363-373.

Skaggs, R.W., G.J. Kriz and R. Bernal, 1973. Field
evaluation of transient drain spacing equation. Trans.
Amer. Soc. Agri. Eng., Vol. 16(3): 590-595.

Skaggs, R.W. and Y.K. Tang. 1979. Effects of drain
diameter openings and evelopes on water table drawdown.
Trans. Amer. Soc. Agri. Eng., Vol. 22(2):326-333.

Smith, M.C. 1983. Subirrigation system control for
water use efficiency. Masters Thesis. Dept. of
Biological and Agricultural Engineering, N.C. State
University, Raleigh, N.C.

Soil Association Map of Michigan. Michigan State
University Extension Bulletin E-lSSO. East
Lansing, Michigan. Dec. 1981.

155

Strickland, E.E., J.T. Ligon, C.W. Doty, and T.V. Wilson.
1981. Water and energy requirements for subsurface
and center-pivot irrigation. Amer. Soc. Agri. Eng.,
Paper No. 81-2069, ASAE, St. Joseph, Mi 49085.

Tang, Y.K. and R.W. Skaggs. 1977. Experimental
evaluation of theoretical solutions for
subsurface drainage and irrigation. The
American Geophysical Union, Paper No. 7W0687,
Vol. 13(6):957-965.

Tisdall, A.L. 1942. Drainage investigations in the
horticultural soils of the Murry Valley. Pamphlet No.
113, Council for Sci. and Ind. Res., Melbourne, Aust.

Unger, P.W. and Danielson, R.E. 1965. Influence of
oxygen and carbon dioxide on germination and seedlings
development of corn (Zea mays L.). Agron. J. 57:50-58.

U.S.D.A. 1954. Diagnosis and Improvement of Saline and
Alkali Soils. AgricuIture Handbook No. 60, Feb.

 

U.S.D.A. 1980. Soil and Water Resources Conversation
Act. Appraisal Review Draft, Part 1.

U.S.D.A. 1980. Soil Survey, Huron county, Michigan,
pp. 66.

U.S.D.A. 1980. Soil Survey, Monroe county, Michigan,
pp. 62.

U.S.D.A. 1984. Soil Survey, Tuscola county, Michigan,
pp. 69-73; Table 18, page 173.

Vissir, W.C. 1958. Delandouwwaterhuishouding in
Netherlands. Comm. 0nderz. Landb. Waterhuish.
Ned. Tno. Rapport nr. 1. 231p.

Vitosh, M. 1984. Irrigation scheduling for potatoes
in Michigan. American Potatoe Journal, Vol. 61:

205-214.

Vomicil, J.A. and Black et.al.(ed.). 1965. Methods of
Soil Anslysis: Part I. Porosity. Agronomy, 9:299-
314.

Wesseling, J. 1974. Crop growth and wet soils. Institute
for Land and Water Management Research, Wageninger,
Netherlands, pp. 7-37.

Williamson, R.E. 1964. The effect of root aeration on
plant growth. Soil Sci. Soc. Am. Proc., 28:86-90.

156

Williamson, R.E. and J.R. Carrekar. 1970. Effect of
water table levels on evapotranspiration and crop
yield. Amer. Soc. Agr., Vol. l3:l68-70,l76.

Williamson, R.E. and G.J. Kris. 1970. Response of
agricultural crops to flooding, depth of water table
and soil gaseous composition. Trans. Amer. Soc.

Agr. Eng., Vol. 13(1):216-220.

Williamson, R.E. and J. van Schilfgaarde. 1965.
Studies of crop response to drainage. II. Lysimeters.
Trans. Amer. Soc. Agr. Eng., Vol. 8:98-101.

Williamson, R.E. and C.R. Willey. 1964. Effect of
depth of water table on yield of tall fescue. Agron.
J. 56:585-588.

Williamson, R.E., C.R. Willy, and T.N. Gray. 1969.
Effect of water table depth and flooding on yield of
millet. Agron. J. 61:844-847.

Wooley, J.T. 1965. Drainage requirement of plants.
Drainage for Efficient Crop Production Conf., Amer.
Soc. Agr. Eng., Proc., Chicago, I11.:1-5.

Worm, B.G., J.T. Ligon, T.V. Wilson, C.W. Doty, and E.E.
Strickland. 1982. Economic evaluation of
subsurface and center pivot irrigation systems.
Proc. of the Specialty Conference on Environmentally
Sound Water and Soil Management. ASCE, Orlando, FL.

nIcnIan STATE UNIV. LIBRARIES
llHIWNIWIIHII”WNWWW||1|1|1||1||H||W|
31293007026754