Crop growth and water use in relation to water management of well-drained sands

MISSING IMAGE

Material Information

Title:
Crop growth and water use in relation to water management of well-drained sands
Physical Description:
xiii, 153 leaves : ill. ; 28 cm.
Language:
English
Creator:
Riestra-Diaz, David, 1948-
Publication Date:

Subjects

Subjects / Keywords:
Soil Science thesis, Ph. D
Dissertations, Academic -- Soil Science -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1984.
Bibliography:
Includes bibliographical references (leaves 148-152).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by David Riestra-Diaz.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
aleph - 020490230
oclc - 30754793
System ID:
AA00026357:00001


This item has the following downloads:


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Tables
        Page vi
        Page vii
        Page viii
    List of Figures
        Page ix
        Page x
        Page xi
    Abstract
        Page xii
        Page xiii
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
    Chapter 2. Literature review
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
    Chapter 3. Materials and methods
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
    Chapter 4. Results and discussion
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
    Chapter 5. General discussion
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
    Chapter 6. Conclusions
        Page 131
        Page 132
    Appendix
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
    Literature cited
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
    Biographical sketch
        Page 153
        Page 154
        Page 155
Full Text











CROP GROWTH AND WATER USE IN RELATION TO
WATER MANAGEMENT OF WELL-DRAINED SANDS















BY

DAVID RIESTRA-DIAZ




















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1984

































To my parents, David and Gabina;

my wife, Maria Guadalupe,

and

daughters,

Nancy, Monica, and Faby,

with love and gratitude













ACKNOWLEDGMENTS


The author is indebted to Dr. L. C. Hammond, chairman of the

supervisory committee, for his support, friendship, encouragement,

and opportunities given during his training program. He gratefully

thanks Drs. P. S. C. Rao, A. G. Smajstrla, K. J. Boote, and J. M.

Bennett, for their frequent assistance. He is also indebted to Messrs.

Alex Clem, Charles Browning, and David Cantlin, for their invaluable

technical assistance and friendship.

Special thanks are given to Mr. Kamarudin Ambak, former graduate

fellow, for his helpful assistance with field work.

The author wishes to express his gratitude to Consejo Nacional

de Ciencia y Tecnologia (CONACYT) of Mexico for granting the scholar-

ship which enabled him to conduct his studies at the University of

Florida.

He is indebted to Dr. Enrique Palacios Velez for cooperation

and frequent aid given to the author during his studies at the Univer-

sity of Florida.

Appreciation is extended to M. C. Gaudencio Diaz Jimenez for his

immense contribution to the author's education, and for his introduction

of the author to the beauty and challenge of water management in agri-

culture.

Most of all, the author wishes to express his eternal appreciation

to his wife, Maria Guadalupe, for all the support, encouragement,

understanding, and help provided throughout these long years of hard

work, especially during periods of stress and uncertainty.



iii














TABLE OF CONTENTS

PAGE

ACKNOWLEDGMENTS..................... ..................... iii

LIST OF TABLES............................................ vi

LIST OF FIGURES........................................... ix

ABSTRACT.................................................. xii

INTRODUCTION.............................................. 1

LITERATURE REVIEW.. ..... ...................... ......... 4

Crop Growth and Water Use............................. 4

Water Management Strategies........................... 9

Water Use Efficiency.................................. 14

MATERIALS AND METHODS ... ................................. 18

General............................................... 18

Oat Lysimeter Experiment.............................. 22

Corn Experiment ...................................... 23

Peanut Experiment ..................................... 25

Soybean Experiment................................... 26

Sweet Potato Experiment............................... 27

RESULTS AND DISCUSSION.................. .................. 30

Oat Lysimeter Experiment.............................. 30

Corn Experiment....................... ................. 44

Peanut Experiment...... ................................ 70

Soybean Experiment................................ ..... 82

Sweet Potato Experiment............................... 104




iv













PAGE

GENERAL DISCUSSION......................................... 124

Crop-Water Production Functions....................... 124

Water Management Strategy............................. 126

Computer-Aided Irrigation Scheduling.................. 127

Future Research Needs................................. 129

CONCLUSIONS............................................... 131

APPENDIX...................................................... 134

LITERATURE CITED ............................................. 148

BIOGRAPHICAL SKETCH....................................... 153









































v














LIST OF TABLES

TABLE PAGE

1 Irrigation schedule, 'Florida 501' oat, 1981-1982.. 32

2 Effect of water management on dry matter yield of
'Florida 501' oat, 1981-1982....................... 36

3 Periodic measured water balance during the vegeta-
tive growth period of 'Florida' 501' oat, 1981-1982 41

4 Total seasonal water balance of 'Florida 501' oat,
1981-1982.......................................... 43

5 Irrigation schedule, 'McCurdy 84aa' corn, 1982..... 46

6 Periodic water balance during the growth period of
'McCurdy 84aa' corn, 1982.......................... 58

7 Total seasonal water balance of 'McCurdy 84aa'
corn, 1982.............................. ........... 64

8 'McCurdy 84aa' corn irrigation-use efficiency,
1982................................................ 69

9 Irrigation schedule, peanuts, 1982.................. 72

10 Peanut pod yields and simulated evapotranspiration
as affected by water management, 1982............... 74

11 Periodic water balance during the growth period of
ten peanut genotypes, 1982......................... 76

12 Total seasonal water balance, peanut genotypes, 1982 79

13 Irrigation schedule, 'Cobb' soybeans, 1982......... 84

14 Periodic water balance during the growth period of
nondefoliated 'Cobb' soybeans, 1982................. 95

15 Total seasonal water balance of nondefoliated
'Cobb' soybeans, 1982.............................. 99

16 Irrigation-use efficiency for 'Cobb' soybeans, 1982 103

17 Irrigation schedule, 'Georgia Jet' and 'Yellow
Jewel' sweet potato, 1982.......................... 106



vi












TABLE PAGE

18 Yields of 'Georgia Jet' (GJ) and 'Yellow Jewel'
(YJ) sweet potato cultivars in response to irri-
gation treatments, 1982............................ 107

19 Periodic water balance during the growth period of
'Georgia Jet' sweet potato, 1982................... 116

20 Total seasonal water balance of 'Georgia Jet' sweet
potato, 1982...................................... 119

21 'Georgia Jet' and 'Yellow Jewel' sweet potato irri-
gation-use efficiency, 1982........................ 123

22 Comparative "measured" and estimated daily ETa as
well as daily ETp, under the criteria of negligible
drainage, full crop canopy, no plant water stress,
and best subplot management conditions............. 128

23 Crop coefficients (Kc) used in the corn experiment. 128

24 Crop coefficients (Kc) used in the peanut experi-
ment ........................... ................... 136

25 Crop coefficients (Kc) used in the soybean experi-
ment.............................................. 137

26 Crop coefficients (Kc) used in the sweet potato
experiment. ....................................... 138

27 Daily potential evapotranspiration rates as
estimated by Penman's method, Gainesville, Florida,
November 1981-November 1982........................ 139

28 Measured and simulated profile water depths with
time for three water management treatments on corn
under nitrogen stress conditions................... 141

29 Measured and simulated profile water depths with
time for four water management treatments on 'Cobb'
soybeans under nondefoliation conditions........... 142

30 Measured and simulated profile water depths with
time for four water management treatments on
'Georgia Jet' sweet potato......................... 143

31 Soil water parameters, Kendrick fine sand, corn
experiment.......................................... 144



vii











TABLE PAGE

32 Soil water parameters, Kendrick fine sand, peanut
experiment. ....................................... 145

33 Soil water parameters, Lake fine sand, soybean
experiment. ....................................... 146

34 Soil water parameters, Lake fine sand, sweet
potato experiment.................................. 147


















































viii














LIST OF FIGURES


FIGURE PAGE

1 Rainfall distribution and irrigation schedules for
'Florida 501' oat dry matter production............. 31

2 Measured average daily ETa, for two treatments as
compared with daily ETp, 'Florida 501' oat, 1981-1982 33

3 Crop water-use coefficients for oat dry matter pro-
duction ............................................ 35

4 Dry matter yield of 'Florida 501' oat versus
seasonal ET and seasonal irrigation (IRRIG).......... 38

5 Relationship between oat relative dry matter yield
decrease (1 Ya/Ym) and the relative seasonal ET
deficit (1 ETa/ETm) ................................ 40

6 The distribution and amount of rainfall and irrigation
during the growing season of 'McCurdy 84aa' corn,
1982................................................. 45

7 'McCurdy 84aa' corn dry matter yield versus seasonal
ET and seasonal irrigation (IRRIG), under nitrogen
sufficient conditions, 1982. (All corn yield data--
Figs. 7 through 12--courtesy of Drs. J. M. Bennett,
L. C. Hammond, J. W. Jones, P. S. C. Rao, and Mr.
L. Mutti).......................................... 48

8 'McCurdy 84aa' corn dry matter yield versus seasonal
ET and seasonal irrigation (IRRIG), under nitrogen
stress conditions, 1982.......... ................... 49

9 'McCurdy 84aa' corn grain yield versus seasonal ET
and seasonal irrigation (IRRIG) under nitrogen
sufficient conditions, 1982......................... 50

10 'McCurdy 84aa' corn grain yield versus seasonal ET
and seasonal irrigation (IRRIG) under nitrogen
stress conditions, 1982............................. 51

11 'McCurdy 84aa' corn evapotranspiration and irrigation
production function under nitrogen sufficient (N )
and nitrogen stress (N ) conditions, 1982........... 53



ix











FIGURE PAGE

12 Relative corn grain yield reduction (1 Ya/Ym)
versus relative ET deficit (1 ETa/ETm) under
nitrogen sufficient conditions..................... 54

13 Root length density distribution of 'McCurdy 84aa'
corn on day 103 after planting..................... 56

14 Measured and simulated root zone water content
distribution with time for the three water manage-
ment treatments, under nitrogen sufficient condi-
tions. 'McCurdy 84aa' corn, 1982.................. 67

15 The distribution and amount of rainfall and irri-
gation during the growing season of peanuts, 1982.. 71

16 Measured and simulated root zone water content dis-
tributions with time under rainfed (treatment 1),
well-irrigated (treatment 2), and irrigated but
induced dry cycle (treatment 3) conditions, peanut
genotypes, 1982......................... ........... 81

17 Distribution and amount of rainfall and irrigation
during the growing season of 'Cobb' soybeans, 1982. 83

18 Grain yield (Y) response of 'Cobb' soybeans to
seasonal amounts of irrigation. (Seed yield data
of Figs. 18 through 21 are courtesy of Drs. K. J.
Boote, J. M. Bennett, and L. C. Hammond)............ 85

19 Grain yield response of 'Cobb' soybeans to seasonal
ETa......................................... ....... 87

20 Relative grain yield (Ya/Ym) of 'Cobb' soybeans
versus relative seasonal evapotranspiration (ETa/
ETm)........................... ....................... 89

21 Relationship between relative grain yield decrease
(1 Ya/Ym) and relative seasonal evapotranspira-
tion deficit (1 ETa/ETm) for 'Cobb' soybeans,
1982 ............................................... 91

22 Relationship between relative yield reduction
(1 Ya/Ym) and relative seasonal evapotranspiration
deficit (1 ETa/ETm) for soybean production at
Gainesville, Florida...................... ........ 93

23 Measured and simulated root zone water content
distribution with time for treatment 1 (rainfed)
and treatment 6 (well-irrigated), nondefoliated
'Cobb' soybeans, 1982 .............................. 101



x












FIGURE PAGE

24 The distribution and amount of rainfall and irriga-
tion during the growing season of 'Georgia Jet'
and 'Yellow Jewel' sweet potato, 1982................ 105

25 Yields of 'Georgia Jet' and 'Yellow Jewel' sweet
potatoes versus seasonal irrigation (IRRIG)
amounts, 1982....................................... 110

26 Yields of 'Georgia Jet' and 'Yellow Jewel' sweet
potatoes versus seasonal ETa, 1982.................. 111

27 Relationship between the realtive sweet potatoes
fresh weight yield decrease and the seasonal rela-
tive ET deficit for the 'Georgia Jet' variety....... 113

28 Average root length density distribution of 'Georgia
Jet' and 'Yellow Jewel' sweet potatoes at harvest
time, 1982.......................................... 115

29 Measured and simulated root zone water content dis-
tribution with time under rainfed (treatment 1) and
well-irrigated (treatment 2) conditions, 'Georgia
Jet' sweet potatoes, 1982 ........................... 121

30 Root depth functions used in the simulation model... 134




























xi














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


CROP GROWTH AND WATER USE IN RELATION TO
WATER MANAGEMENT OF WELL-DRAINED SANDS


BY

DAVID RIESTRA-DIAZ

April 1984

Chairman: Dr. L. C. Hammond
Major Department: Soil Science

Field experiments were conducted to develop water production

functions for oats, corn, peanuts, soybeans, and sweet potatoes.

Irrigation strategy was to apply small amounts of water to only partially

replenish the soil water deficit in the root zone in order to minimize

transport of water and nutrients beyond the crop root zone. Simulation

model (NITROSIM) was used to estimate evapotranspiration and drainage,

and soil water-storage within the root zone.

For all crops, yields increased linearly with seasonal irrigation

and evapotranspiration. Crop-water production functions for oat dry

-1 -1
matter were 556 and 499 kg ha cm respectively, for actual evapo-

transpiration (ETa) and irrigation. Due to limited data the only pro-

duction function for peanuts was pod yield versus ETa (162kg ha- cm-).

In the soybean study, there were two ETa production functions

which reflected differences in timing of ETa deficits. Irrigating

during pod set and pod filling stages resulted in an ETa production
-1 -1
function of 75 kg ha cm Witholding irrigation during these



xii










-1 -1
critical growth stages produced an ETa function of 132 kg ha cm

A unit ETa deficit during pod set and pod filling produced about 1.8

times more yield decrease than a unit ETa deficit during vegetative

growth. 'Georgia Jet' sweet potato responded to water management

while 'Yellow Jewel' did not. The irrigation and ETa production

functions for marketable yield of 'Georgia Jet' were 2348 and 3188
-1 -1
kg ha cm respectively.

Our water management strategy gave variable results depending

on the combination of rainfall events and irrigation inputs. Water-

use efficiencies (percent of irrigation used to increase ETa) varied from

a low of 34% in the peanut experiment to a high of 89% in the oat experi-

ment. In corn, soybeans, and sweet potatoes, the respective water-use

efficiencies were 75, 53, and 51% for the better subtreatments.

The simulation model (NITROSIM) provided reasonable estimates of

ETa as well as soil-water storage in the root zone. The model will

be a useful research tool for the evaluation of water management prac-

tices under Florida soil and climate conditions.





















xiii













INTRODUCTION


The vital role of water for food production is well known in both

rainfed and irrigated agriculture. In rainfed agriculture, poor sea-

sonal distribution of precipitation and large variations from year to

year may cause complete crop failure. In many humid regions, drought

periods are especially detrimental on sandy soils and with crops that

have shallow root systems. Hence, there has been a major growth in

irrigated agriculture in many rainfed regions of the world. Irrigated

agriculture, at present, is entering an age of management in which water

deficits in crop production may not be totally avoided, but instead

favorably controlled (Garrity et al., 1982a).

There are unique differences in water management needs between

arid and humid regions. In arid regions, irrigation is required to

supply nearly all the crop water needs plus an additional non-evapo-

transpirative fraction of water to maintain a favorable salt balance

in the root zone. By contrast, in humid regions, irrigation supplies

less than 50% of the crop seasonal water needs and little if any irri-

gation is needed for maintaining salt balance. The goal of water

management (irrigation) in humid regions is to maximize evapotranspira-

tive use of the irrigation fraction of the total water input while

minimizing loss of water, fertilizers, and pesticides from deep perco-

lation beyond the root zone. Research efforts to attain this goal are

only in the beginning stages. The development of irrigation management

techniques which are necessary for efficient and environmentally sound

uses of water, fertilizer, and pesticides requires additional research.



1






2



More information is needed on irrigation scheduling (timing, frequency,

and quantity per irrigation) as well as on the design and operation of

irrigation systems compatible with a wide range of crop, soil, and

climate requirements. The whole physical-biological system must be

studied to provide basic information in the following areas: metero-

logical and crop factors of evapotranspiration, system water balance,

plant response to water stress at different growth stages, plant and

soil factors influencing root system development, and root density

distribution; uptake of water and nutrients, plant water stress in

relation to water availability and temporal and spatial variability

of soil matric potential; water and solute retention and transport in

soil and plants, and drought avoidance mechanisms in plants. Irrigation

strategies must be developed and tested. Supporting information should

include weather forecasting, crop water production functions, risk and

economic analyses, interactions of plant stresses (water, nutrients,

temperature, and biological), and factors of water-use efficiency.

In addition to crop production aspects of water management, there

is concern about water conservation. Even in humid regions like Florida

with abundant but finite water resources, state water management dis-

tricts and other regulating authorities are concerned with protecting

surface and groundwater quantity and quality.

The present study was undertaken with the following objectives:

(i) To develop evapotranspiration and irrigation production func-

tions for oats, soybeans, peanuts, corn, and sweet potatoes under

Florida climate and soil conditions.

(ii) To develop economically beneficial irrigation scheduling







3


strategies based on wetting depth, timing of irrigation, weather fore-

casting and crop growth stage sensitivity to water stress. The basic

strategy is to replenish a part rather than the full soil water deficit

in the root zone in order to maximize evapotranspirative use of water

and minimize transport of water, nutrients, and pesticides beyond the

root zone.

(iii) To measure the water balance consequences of different

water management strategies, and to use an existing simulation model

(NITROSIM) to predict daily and seasonal evapotranspiration, drainage,

and soil-water storage within the root zone.














LITERATURE REVIEW


Crop Growth and Water Use


There is experimental evidence that, for many crops, plant growth

in terms of dry matter yield is proportional to the amount of transpi-

ration (DeWit, 1958; Arkley, 1963; Hillel and Guron, 1973; Tanner and

Sinclair, 1983). Monteith (1965) indicated that the close correlation

between transpiration and dry matter yield could be explained by the

fact that net radiation, which determines to a large extent the trans-

piration rate, and solar radiation, which determines photosynthesis,

are linearly related. The stomata are the valves which allow CO2 to

enter the leaf to be available for photosynthesis. However, at the

same time, water loss (transpiration) occurs through the open stomates.

Water stress results in a closing of stomates and a reduction in both

transpiration and CO2 exchange. Since the same stomatal barrier is

encountered by both CO2 and water vapor during photosynthesis and

transpiration, a linear relationship between photosynthesis and trans-

piration would be expected (Sinclair et al., 1984).

Bierhuizen and Slatyer (1965) explain the photosynthesis-trans-

piration relationship in terms of the vapor pressure deficit of the air.

The processes are described in the following diffusion equations:


a Ae
T (1)
P r +r (
a a s

ACO
P G (2)
P r' +r' +r '
a s m


4







5


2 -1
where T is the transpiration rate (kg m s ); Ae is the vapor pressure

deficit beween leaf and air; r and r are the water vapor diffusion
a s
resistances resulting from the leaf laminar boundary layer and stomata;

E is the ratio of the mole weight of water vapor to air; p and P are
a a
the air density and atmospheric pressure; P is the photosynthesis rate
-2 -1
(kg m s ); ACO2 is the difference in carbon dioxide concentration

of the atmosphere and at the CO2 fixation site; r' and r' are the
2 a s
boundary layer and stomatal resistances to CO2 diffusion into the leaf;

and rm describes the CO2 diffusion resistance into the cells to the

chloroplasts. These research workers then expressed the photosynthesis

transpiration relationship as:



T EPa r' +r' +r'
T = a a s m 1
-Ae (3)
P P r +r ACO e. (3)
a a s 2


Sinclair et al. (1984) stated that equation 1 suggests that

plant transpiration rate is proportional to the vapor pressure dif-

ference between the inside of the leaf (saturated vapor pressure at

the leaf temperature) and the bulk air.

Arkley (1963) pointed out that the soil evaporation component of

evapotranspirative water use contributes little to plant growth.

Transpiration, on the other hand, is directly involved in the growth

of nearly all higher plants. Thus, Arkley found that many experiments

in the literature showed a highly correlated linear relationship between

dry matter yield and transpiration. Plants were grown in containers

where soil evaporation was prevented, and plant transpiration was

corrected for mean relative humidity during the period of most active

growth.








6


Under field conditions, crop water use is usually expressed as

evapotranspiration (ET) due to the fact that both components of ET,

evaporation and transpiration, are simultaneous processes, and soil

evaporation is not practically distinguishable or measured. Thus,

many studies have shown that dry matter of field crops is linearly

related to ET (Arkley, 1963; Hanks et al., 1969; Hillel and Guron,

1973; Stewart et al., 1977; Verasan and Phillips, 1978; Fischer, 1979).

Numerous field experiments have been conducted to study the rela-

tionship between marketable crop yields and cumulative seasonal ET.

In general, the relationships have been linear (Jensen and Musick,

1960; Musick et al., 1963; Stewart et al., 1977; Skogerboe et al.,

1979; Hammond et al., 1981b). However, some researchers have reported

a curvilinear relationship and fitted quadratic expressions to the

results (Musick et al., 1976; Palacios, 1977).

From the point of view of water management it is important to know

under what circumstances the linear and curvilinear relationships apply.

Timing of water deficits has been found to be a major cause of yield-ET

variations from the linear relationship. Water deficits occurring at

one growth stage might have a different effect on crop yield as compared

with deficits at another stage. Thus, ET deficits during more sensitive

or "critical" crop growth stages cause a relatively larger decrease in

yield. Stewart et al. (1976) pointed out that when the timing of ET

deficit is optimal (i.e., the minimum yield loss to be expected from

any given seasonal ET deficit) the relationship bewteen yield and

seasonal ET is represented by a straight line function. Sinclair et al.

(1984) stated that the linearity of the response could be explained by








7


the fact that in most of the experimental data the crop harvest index

(defined as the ratio of grain dry matter to the total dry matter

yield) showed little variation over a moderate range of evapotrans-

piration deficits.

Stegman et al. (1980) reported that several researchers have found

a relative increase in cell solute concentration during the gradual

development of stress in field grown plants. With this solute increase

a more negative leaf water potential develops before stomata close, and

plants can tolerate more water stress before photosynthesis is drasti-

cally reduced.

Curvilinear relationships between grain yield and seasonal irriga-

tion are mainly due to the combination of two causes (Barrett and

Skogerboe, 1980). Firstly, when water is applied in excess of the

amount required for maximum evapotranspiration (ETm) and maximum yield

(Ym), crop yields remain constant or decrease with increased seasonal

irrigation amounts. Secondly, in most cases, scientists have not

reported yields with ET as the independent variable, but rather water

supplied by irrigation. The yield-irrigation relationship is very

sensitive to non-ET losses. When Hillel and Guron (1973) measured

the drainage component of the soil water balance, the relationship

between corn grain yield and seasonal ET was strongly linear.

Stewart and Hagan (1973) have well illustrated the difference in

the crop yield response curves obtained with seasonal ET versus seasonal

irrigation. Stegman et al. (1980) stated that the non-ET losses were

due to inefficiencies in the irrigation method as well as the inexact-

ness involved in scheduling irrigation according to plans.







8



Other researchers have found that when irrigation was applied in

amounts nearly equal to the maximum crop evapotranspiration needs,

there was a linear relationship between yield and irrigation. Such

linear relationships have been presented for corn (Hammond et al.,

1981b), sorghum (Garrity et al., 1982a, 1982b), and soybeans (Hammond

et al., 1981b; Ambak, 1982). Thus, if all of the water applied was

used for the crop ET, the irrigation and ET functions would be the

same. However, the irrigation function always has a smaller slope

than does the ET function. The ratio of the two slopes is a measure

of irrigation-use efficiency for the specific study (Stewart and Hagan,

1973; Stegman et al., 1980; Hammond et al., 1981a). Hammond et al.

(1981a) pointed out that the irrigation function must be determined

through research in order to develop economic water management systems

and practices for particular soil-climate-crop-systems.

As a concluding remark from the above review of plant growth and

water use, it is obvious that crop production is linked to crop trans-

piration. To increase crop biomass and grain production more transpi-

rative water needs to be used in agricultural production systems. When

other growth factors are non-limiting and water amounts are adequately

supplied during the crop season, the nature of the functional relation-

ship between crop production and water use in terms of transpiration,

evapotranspiration or irrigation is basically linear (Tanner and

Sinclair, 1983; Sinclair et al., 1984).







9



Water Management Strategies


Stegman et al. (1980) stated that water management practices are

generally adapted to meet the following objectives: (a) maximizing

yield per unit of land, (b) maximizing yield per unit of water applied,

(c) maximizing net profit, and (d) minimizing energy requirements.

Thus, many researchers have developed conceptual irrigation scheduling

practices which have been based on particular production systems under

arid and humid climatic conditions.

Hiler et al. (1974) developed the stress day index concept (SDI)

for optimum timing of irrigation as follows:


SDI = (SD. x CS.) (4)


where SD. is the stress day factor which expresses the degree of

water deficit in specific plant growth stage i; CS. is the corp sus-
1
ceptibility factor which measures the susceptibility of plant growth

stage i to a given water deficit. The crop susceptibility factor

was defined as:

X X.
CS. = (5)
1
X

where X is the marketable yield produced by a control treatment which

is kept well-watered throughout the season; Xi is the yield in a

treatment that was subjected to deficit only during the ith growth

stage. These researchers used leaf water potential (bars) as a measure

of SDi for sorghum. They found critical SDI (SDIo) values ranging

from 4 to 16 bars depending upon growth stage. The method is dependent

upon experimentatal determinations of both SD. and CS. for each parti-
1 1
cular crop-soil-climate system.








10


Jensen et al. (1970) proposed the use of a water simulation

model with allowable depletion adjusted for the stage of corp growth

plus regular updating with some measure of the soil water content in

the root zone. Earlier, Jensen (1968) developed a multiplicative

type expression to relate the effects of limited soil water on grain

yields:

n .
Ya _= n ETa i
Ym i=l EpD (6)
i

where Ya/Ym is the relative marketable yield; (ETa/ETp) is the relative

total evapotranspiration during growth stage i; X. is the relative
1
sensitivity of the crop to water stress in growth stage i; and n is the

number of growth stages considered. The implication of this model is

that the interactive effect of water stress at various growth stages

on crop yield may result in some non-linearity between yield and sea-

sonal evapotranspiration. However, limited water resources can be

properly allocated by only irrigating during growth stages where yield

is sensitive to water stress. Recently, Smajstrla and Clark (1981)

determined Ai values for Williams soybeans grown as early season soy-

beans in Florida.

A plow-layer soil water management and programmed fertilization

system has been developed by Rhoads (1981) for Florida Ultisols. The

system involves the application of enough water to recharge the soil

to plow depth before yield limiting water stress develops, and the

periodic addition of plant nutrients at rates to meet the needs of

plants growing at maximum rates. Rhoads obtained highest corn yields

by irrigating when the soil-water tension reached -20 kPa in the plow








11



layer. According to Rhoads, multiple applications of fertilizer

coupled with the plow-layer water management strategy is the most

practical approach to protecting fertilizer from leaching. Neverthe-

less, Rhoad's system might be made more efficient with due considera-

tion of the drainage consequences of the method.

Phene and Beale (1976) proposed a water and nutrient managment

method in which the optimal range of soil matric potential was based

on soil oxygen diffusion rate, soil strength, water desorption charac-

teristics and unsaturated hydraulic conductivity. A high-frequency

irrigation strategy was used in their studies on sweet corn. These

researchers pointed out that nitrogen leaching can be minimized when

soluble nitrogen and potassium are applied frequently through the

irrigation system. Again, the question of drainage loss as a conse-

quence of a nearly constant condition of maximum available water

capacity is not addressed.

Rawlins and Raats (1975) discussed irrigation scheduling strate-

gies that deal with gradual depletion of available soil water during

the growing season. They indicated that high-frequency irrigation goes

a long way toward meeting the conflicting requirements of maintaining

a high plant water potential and a sufficient capacity to store erratic

rainfall.

Hammond et al. (1981a) pointed out that timing, application inten-

sity, method of application and amounts of water applied affect the

fraction of added water distributed to the depletion components (ET

and drainage from the root zone) and to soil water storage. Leaching

losses of fertilizer and pesticides occur when there is drainage, and









12


in some cases, the aeration condition of the soil root zone may be

affected unfavorably during high soil water conditions. Fertilizer

and water management should be coupled, especially in sandy soils

where irrigation and rainfall are potentially responsible for the

leaching of highly mobile ions such as NO3 (Burns, 1980). This

problem has received limited research attention. Recently, Tanner

and Sinclair (1983) concluded that fertility management practices in

humid regions have developed with little regard to the interaction

between nitrogen-use efficiency and irrigation efficiency. Thus,

considerable research must be done to develop sound fertilizer/

irrigation practices which minimize drainage and leaching while main-

taining an adequate water and nutrient supply for the crop.

The stress degree day (SDD) concept was developed by Jackson et

al. (1977) for timing irrigation. The SDD is based on the difference

between the temperature of a plant canopy (Tc) and the temperature

of the surrounding air (T ). It has been established that as water
a
becomes limiting, T relative to T increases due to lack of water
a c
for transpiration. Thus, SDD was defined as:


N
SDD = LCT T n (7)


where (T T ) is summed over N days beginning at day i. In general,
c a
if a plant is well irrigated, T T will be near zero or negative.
c a
If it is water stressed, T T will be positive. The sum of positive
c a
values will be an index of when to irrigate, i.e., water will be

applied when the SDD reaches a critical value. Jackson et al. (1977)

used a critival value of 10 in their studies.







13




Geiser et al. (1982) developed an irrigation scheduling model by

using crop-canopy-air temperature difference (AT) as the dependent

variable and net radiation (Rn), relative humidity and available soil

water (ASW) as independent variables. The model developed from nulti-

ple regression analysis was:


AT = -1.065+4.71x103Rn+0.027RH-0.0535ASW+4.01x10-10ASW5 (8)


In order to use the model for irrigation scheduling, the variable ASW

was set as a constant (50% depletion), and AT was calculated for various

conditions of net radiation and relative humidities. Thus, the AT

value obtained from a particular Rn and RH condition represented a

critical AT value (ATc) at which irrigation was scheduled. As an

example of the procedure, these researchers indicated that with a net

radiation of 600 W/M2 and relative humidity of 50%, a critical value

of ATc of 0.575 C was determined.

As a summary of the above section, the SDI irrigation timing

scheme developed by Hiler et al. (1974) follows the objective of

maximizing yields per unit of water applied, especially in regions

where water is a limiting resource. Both components SD. and CS.
1 1

need to be determined from research for each particular crop.

Jensen's model follows the objective of increasing yield per unit

of water applied. However, this model has limited usefulness because

of the general unavailability of the sensitivity coefficients.

The plow-layer soil management and program fertilization concept

developed by Rhoads (1981) for Florida Ultisols, maximizes crop yield

per unit area under the assumption of an unlimited water resource.







14


Even though fertilization is applied frequently to avoid leaching

losses, the possibility always exists that water and fertilizer will

be displaced when unexpected heavy rainfall events occur after the

plow layer has been totally replenished by irrigation.

The water-nutrient management method developed by Phene and Beale

(1976) for humid regions maximizes crop yield and net profit per unit

of water applied, and minimizes energy requirements. Because the

system is based on high-frequency irrigation with a totally replenished

soil profile, there is a high risk in humid regions of water loss by

drainage and ground-water contamination with nitrate.

The stress day degree method developed by Jackson et al. (1977)

is based on a canopy temperature indicator. Although the feasibility

of using thermal measurements to evaluate water stress in plants has

been demonstrated, few producers have instrumentation available for

timing irrigation. This method may have utility in humid regions

if the effect of vapor pressure deficits and solar radiation are

considered (Geiser et al., 1982).


Water Use Efficiency


Water use efficiency (WUE) as defined by Viets (1962) as the

mass ratio of crop yield to water use. Many variations of this basic

definition have appeared in literature. In quantitative terms WUE

has been variously defined as the ratio of biomass accumulation,

expressed as carbon dioxide assimilation, total crop biomass, or grain

yield, to water use expressed as transpiration, evapotranspiration, or

total water input to the system. The time-scale may be instantaneous,







15



daily or seasonal (Sinclair et al., 1984). Water use efficiency can

be increased by (a) increasing yield without increasing water use, or

(b) maintaining equal yield and decreasing water use. Yields can be

increased by use of fertilizer, plant breeding, better methods of pest

and disease control, and improving supplies of sunlight and carbon

dioxide to the leaves. Some of these practices will invariably cause

some change in the water use pattern and will directly affect WUE.

Hillel and Guron (1973) state that it appears more promising to attempt

to increase WUE by increasing crop yields than by decreasing evapotrans-

piration, since plants in the field are subject to an externally imposed

evaporative demand.

Viets (1966) and Black (1966) have shown that WUE is increased by

improved soil fertility and cultural practices. These practices in-

crease dry matter production without increasing ETa. However, Tanner

and Sinclair (1983) and Sinclair et al. (1984) argued that WUE remains

relatively constant for a particular crop regardless of improvements

in soil management and cultural practices.

Recently, Sinclair et al. (1984) derived an expression for seasonal

evapotranspirational water-use efficiency for grain yield, WUE (ET) as

follows:

WUE (ET) = HIfKd ET/(e*) KdE/e--]//ET. (9)
a a
where:

H = harvest index (ratio of grain to total dry matter yield),

Kd = constant incorporating plant characteristics and the ener-

getics of CO2 conversion to biomass,







16




ee = mean daily vapor pressure deficity, weighted only for
e*-e
a
the periods of transpiration,

ET = seasonal evapotranspiration, and

E = seasonal evaporation from the soil.

According to these researchers, the integrals of equation 9 can be

eliminated with the assumption of stable conditions (Kd, E, and (e*-e)
a
are reasonably constant] during the season. Thus, equation 9 was

expressed as:


WUE (ET) = C1 E/ET)HKd/(-e)). (10)
a
Equation 10 shows the effect of soil evaporation on WUE (ET) especially

during the early vegetative growth when plants have not developed full

canopy cover. Maximum transpirational water-use efficiency WUT(T) is

reached when E is reduced to zero. As mentioned by Sinclair et al.

(1984), the WUE (ET) can be improved by management practices that tend

to minimize the ratio E/ET. An example practice is narrow-row spacings

which result in a more rapid development of full crop canopy.

Additional mecanisms that influence water-use efficiency at

different scales of observation have been delineated recently by

Tanner and Sinclair (1983) and Sinclair et al. (1984). These researchers

gave several viable options for improving WUE such as biochemical alter-

ations, stomatal manipulation, alteration of cropping seasons, improved

harvest index and increased plant transpiration (avoiding water deficits).

From the above review, one can conclude that maximum yields and

high water-use efficiencies occur at the point where seasonal water use

equals the evaporative demand of the atmosphere. Scheduling irrigation






17



strategy plays an important role in achieving this goal, especially

in humid regions where water resources are available and irrigation

substantially increases agricultural production. Tanner and Sinclair

pointed out that "if we are to increase national food production,

the greatest increase per investment is likely to derive where produc-

tivity already is high but limited by management rather than resource"

(1983, p. 24 ). And further, the authors indicated that "we might best

make expenditures to learn how to manage water well in the humid re-

gions where there is water" (p. 24).













MATERIALS AND METHODS


General


From the period November 1981 to November 1982, five field

water management experiments were conducted at the Irrigation Research

and Education Park (IREP) on the University of Florida campus, Gaines-

ville. Each experiment involved one of the following crops: corn,

soybean, peanuts, oats, and sweet potatoes.

Corn and peanut experiments were established on a Kendrick fine

sand (loamy siliceous, hyperthermic family of Arenic Paleudults) which

consists of fine sand material over a fine sandy loam B2t horizon be-

ginning at a depth of 100 to 150 cm. The oat, soybean, and sweet

potato experiments were established on a Lake fine sand (hyperthermic

coated family of Typic Quartzipsamments).

Water management treatments (rainfed and various seasonal irriga-

tion levels) were designed to obtain a series of seasonal ET amounts

in each experiment. The basic irrigation strategy was one of supplying

water frequently and in amounts required to refill only a fraction of

the water depleted in the root zone. Different levels of seasonal irri-

gation were obtained by witholding irrigation during certain droughts,

varying irrigation timing and amounts, and by using portable rain shel-

ters at selected growth stages. In the corn, soybean, and peanut experi-

ments, water was applied at the rate of 2.54 cm from a solid-set impact

sprinkler system. Quarter circle sprinklers located at the corners

of 14m x 14 m plots gave a full two-sprinkler overlap along the plot



18





19



edges. All four replications of a particular treatment were irri-

gated at the same time.

In the oat lysimeter experiment measured amounts of water were

applied by hand from a watering can.

A portable, low pressure (20 psi) aluminum frame "Microjet" system

delivered water at a rate of 1.47 cm hr-1 to the sweet potato experi-

ment. Plot size was 4.6 m x 6.1 m. Soil water status in all experi-

ments was measured periodically during the growing season using gravi-

metric, neutron scattering and tensiometric methods. Measurements

were made at 15 cm depth increments in two of the four replications.

Neutron readins (quarter-minute counting time) were obtained at least

twice a week with a Troxler" portable scaler and neutron depth probe

(100 mCi Am-Be) model 1651. Gravimetric soil samples were taken in-

frequently only for special needs such as a more precise measurement

of the water content in the 0-30 cm soil layer. Hydraulic head (H)

and matric suction (h) were measured with sets of tensiometers in-

stalled adjacent (30 to 60 cm) to neutron access tubes. Irrigation

timing was based on selected values of soil water suction at 15 cm

depths.

Periodic and seasonal water depletion amounts were calculated

from measured soil water contents, rainfall, and irrigation assuming

no surface runoff and no upward transfer of water into the root zone.

Stated in equation form:


SWD. = R. + I. + AS = ETa. + D. (11)
1 1 1 1 1 1







20



where:

SWD. = soil water depletion for period i (cm),
1
R. = rainfall inputs for period i (cm),
1

I. = irrigation inputs for period i (cm),
1
AS. = change in soil water storage in the root zone for period
1

i (cm),

ETa. = actual evapotranspiration for period i (cm), and
1

D. = drainage loss from the root zone for period i (cm).

The actual evapotranspiration (ETa.) and drainage components of

SWD. as well as soil water contents were calculated with the simulation
1

model (NITROSIM) (Rao et al., 1976, 1981). This model incorporated

estimated daily potential evapotranspiration (ETp), measured soil

water characteristics [field capacity (FC), permanent wilting percentage

(PWP), and water redistribution time], estimated root depth with time,

and a water extraction rate equal to the ETp rate until 40% of the

available water (AW) in the root zone had been depleted. At that point,

the extraction rate was decreased linearly with decreasing AW to zero

at PWP. The Penman equation (Penman, 1948) was used to estimate daily

ETp rates considering an albedo of 0.23 (vegetated surface). Official

NOAA daily weather data supplied by the Department of Agronomy are not

included here.

Certain detailed model parameters are given in the Appendix. Root

depth versus time was crop and site specific. The values used in this

study were based on past experiences of researchers at the IREP and

current observations of rooting depth and soil water extraction patterns.

Crop coefficients were crop and treatment specific. The primary criterion





21



for selection of these coefficients was current observations of the

crop canopy development. This was supplemented with current leaf

area index (corn) and published crop coefficients (Palacios, 1977).

Soil water retention parameters were site specific and measured in

the present study. Permanent wilting point was measured as the labora-

tory determined 15-bar water content, while the field capacity was

measured (neutron and gravimetric) in the field.

It was found that the model required calibration of the water

redistribution parameter in order to give reasonable agreement between

measured and simulated soil water contents. The water redistribution

equation in the NTTROSIM model was:


et = FC + 0S 0FC exp (-at) (12)

where 0 values are the volumetric water contents at time t days, field

capacity, and infiltration content (maximum) for subscripts t, FC, and

S, respectively. The value of a is -Zn 0.01/t.

In the current study, the above equation was found to redistribute

the infiltrated water too rapidly. The equation was modified to


Ot = OR + -S OR exp (-at) (13)


where a and t are 0.167 and 30 days, respectively. The term 0R is a

residual water content related to 0FC and 015 by


OR = U(FC 15)/] + 015 (14)

where 015 is the volumetric water content at 15 bars water suction.

These 0 values are averages for a specific soil profile and the coef-

ficient f is defined in the following function:






22



f (aR 015) =(FC 15) (15)

where the 0 values are average volumetric water contents as in

equation 14.

All modeling and statistical analyses were performed at the North

East Regional Data Center (NERDC) of the State University System of

Florida, Gainesville, Florida. Statistical Analysis Systems (SAS 79.5)

programs were used for analysis of variance, and the Duncan Multiple

Range.Test was used to compare treatment means. Linear regression

analyses were made with a programmable calculator.

Crop response to water management in all experiments was

measured in terms of total dry matter and/or marketable yields.


Oat Lysimeter Experiment


Oats (Avena sativa L.) variety 'Florida 501' were planted in a

35 m x 25 m field containing 20 lysimeters (closed-bottom steel tanks
2
1.67min diameter, 2.2 m cross section, and 1.8 m deep). A porous

ceramic plate in the bottom of the lysimeters provided for the vacuum

pump extraction of drainage water to a suction of 7 kPa (Smajstrla,

1982).

The soil was prepared by a rototill incorporation of a broadcast

application of NH NO3 at a rate of 60 kg ha-1 N immediately before

planting. Oat seeds were planted on 20 Nov. 1981 at a depth of 2.5 cm

with approximately 22 kg of seed per ha. An additional NH NO3 applica-
-i
tion (15 kg ha- N) was made at 46 days post planting.

Five water management treatments were used in a randomized complete

block design replicated four times. The five treatments were:







23


1. Rainfed

2. Irrigation, light-frequent, when soil-water water suction

was 15 kPa at a depth of 15 cm

3. Irrigated, light-infrequent, when soil suction was 30 kPa

at a depth of 15 cm

4. Irrigated, medium-infrequent, when soil water suction was

30 kPa at a depth of 30 cm

5. Irrigated, light to medium-infrequent, when soil water

suction was 30 kPa at a depth of 30 cm.

The above-ground vegetative biomass was harvested from the

lysimeters on 1 Mar. 1982 as heading was beginning. The plants were

dried in a forced-draft oven at 700C for several days before weighing.


Corn Experiment


This experiment was designed and managed by a multidiscipline

team of researchers consisting of Drs. J. M. Bennett (Department of

Agronomy), Dr. J. W. Jones (Department of Agricultural Engineering),

and Drs. L. C. Hammond and P. S. C. Rao (Department of Soil Science).

Yield data were made available by courtesy of them for crop-water

production function analysis only. Detailed yield data will not be

presented here.

'McCurdy 84aa' corn (Zea mays L.) was planted on 26 Feb. 1982 in

field unit 3B, IREP. Prior to planting, the land was plowed to a depth

of 20-25 cm with a moldboard plow. A disk harrow incorporated 1120 kg
-I
ha of a broadcast application of a mixed commercial fertilizer: 0-10-25

(N-P205-K20) plus 5.8% Mg, 0.75% Mn, 0.75% Zn, o.25% B, and 6.1% S

from MgSO4.






24



Corn seeds were planted in 60-cm rows with in-row spacing of 22 cm,

resulting in a plant population of approximately 71,660 plants per ha.

The experiment was a four replicate split-plot design with three

irrigation treatments as main plots and two nitrogen fertilizer rates

as subplots. The water management treatments were as follow:

1. Rainfed

2. Irrigated (optimal) when soil water suction reached 20 kPa

at 15 cm soil depth

3. Irrigated (stress) same as treatment 2 except that a 2 to

3 week period of no irrigation for drought stress was planned

to begin at 50 days.

Nitrogen fertilizer treatments were:

NO: Nitrogen stress during vegetative growth. Sidedress appli-

cation of 27, 36, and 54 kg ha-1 of N at 31, 35, and 67 days,

respectively

N1: Optimal nitrogen. Sidedress applications of 64, 52, 75,
-I
37, 55, and 126 kg ha-1 N at 5, 20, 35, 51, 60, and 67 days,

respectively.
3
Root-length densities (cm of root per cm soil) were measured in

all irrigation and fertilizer treatments of replication four. Measure-

ments were made twice during the season (93 and 103 days). Soil samples

were collected in four sites of each treatment (about 5-10 cm from the

stalk). Samples were collected to a depth of 120 cm in 15 cm increments

with a 5-cm aluminum tube. The soil samples were washed on a sieve and

roots were collected and spread on a piece of nylon mesh superimposed

on a plastic sheet containing 1 x 1 cm grid lines. Root and grid line






25



intersections were counted for calculation of root lengths (Tennant,

1975).

Corn grain was harvested on 7 July 1982, a few days after black

layer development.


Peanut Experiment


This experiment was designed and managed by a multidiscipline team

of researchers consisting of Drs. K. J. Boote and J. M. Bennett,

Department of Agronomy, and Dr. L. C. Hammond, Department of Soil

Science. Yield data were made available by courtesy of them for crop-

water production function analysis only. Detailed yield data will not

be presented here.

Peanut (Arachis hypogea L.) seeds were hand-planted at 10 cm

spacings in 60 cm rows on 5 May 1982 in field unit 3A of the IREP.

Prior to planting, the land was plowed and a broadcast application

(500 kg ha-1) of mixed fertilizer was incorporated with a disk harrow.

The fertilizer contained 6, 25, 6.5, 5.8, 0.75, 0.75, and 0.25% of

P205, K20, Mg, S, Mn, Zn, and B, respectively. Gypsum (CaSO 4 H 0)
-l
was broadcast over the plants at a rate of 1120 kg ha- on 16 June.

The experiment consisted of three water management treatments

(main plots) and 10 peanut genotypes (subplots) arranged in a split-

plot design replicated four times. Main plots (14 m x 14 m) consisted

of a total of 40 rows, with each peanut genotype occupying four rows.

Reported yields will be averaged over the 10 peanut genotypes used in

the experiment: UF 714021, Florigiant, UF 78114, UF 781181-1217, PI

383426, UF 79131-9, Early Bunch, UF 77318, Florunner, and UF 75102.







26


Water management treatments were:

1. Rainfed

2. Irrigated (optimal) when soil water suction reached 20 kPa

at 15 cm soil depth

3. Irrigated (stress) same schedule as treatment 2, except that

water stress was induced on subplots of genotypes UF 781181-

1217, PI 383426, Early Bunch, and Florunner during the

period of 3 July to 2 August. Reported yields for this

treatment will be averaged over the four genotypes.

Water stress in treatment 3 was imposed by covering the subplots

with portable rain shelters during expected rainfall events.

On 20 Sept. 1982, a total of 6 m of row from the two center rows

of each genotype were harvested by hand. Pod yields were measured and
-1
expressed in kg ha1


Soybean Experiment


This experiment was designed and managed by a multidiscipline

team of researchers consisting of Drs. K. J. Boote and J. M. Bennett,

Department of Agronomy, and Dr. L. C. Hammond, Department of Soil

Science. Yield data were made available by courtesy of them for crop-

water production analysis only. Detailed yield data will not be

presented here.

'Cobb' soybeans (Glycine max L.) were planted on 30 June 1982

with 40 seed per m in 76 cm rows. Field unit 2 of IREP was used.

The experiment consisted of a four replicate, split-plot design

where six water management treatments were used as main plots with






27



two induced insect defoliation levels (zero and 40-50%) as subplots.

Water management treatments were:

1. Rainfed

2. Irrigated only in reproductive stages R4 and R6 when soil

water suction at 15 cm soil depth reached 20 kPa

3. Irrigated as in treatment 2, but only between reproductive

stages R4 and R7

4. Irrigated as in treatment 2, except that natural drought

stress was permitted to develop between reproductive stages

R4 and R6

5. Irrigated as in treatment 2, except that natural drought

stress was permitted to develop between reproductive stages

R6 and R7

6. Irrigated as in treatment 2, except that irrigation was

scheduled as needed throughout the season.

Soybean harvest on 4 November consisted of three rows of 3 m

length each where the stand was uniform in the center of each subplot.

Beans were mechanically threshed, cleaned, and weighed. Yields were
-I
expressed in kg ha


Sweet Potato Experiment


Sweet potato (Convolvulus batatas L. cultivars 'Georgia Jet' and

"Yellow Jewel') vine cuttings were planted on 9 July 1982 in rows 35 cm

apart with 30 cm between plants.

Prior to planting the experimental area was broadcast fertilized

with 1120 kg ha- of the fertilizer formulation used in the corn
with 1120 kg ha of the fertilizer formulation used in the corn







28


-i
experiment. Two broadcast applications of 80 and 56 kg ha N in the

form of NH NO3 were made at 17 and 41 days after planting, respectively.

The experiment consisted of a four replicate, split-plot design

with six water management treatments as main plots and two varieties

as subplots. Water management treatments were:

1. Rainfed

2. Irrigated (optimal) when soil water suction reached 20 kPa

at 15 cm soil depth

3. Irrigated, when soil water suction reached 20 kPa at 30 cm

soil depth

4. Irrigated, when plants exhibited visible wilt

5. Irrigated (stressed) same as treatment 2, except that water

stress was induced during the period 40 to 70 days after

planting

6. Irrigated (stressed) same as treatment 2, except that water

stress was induced during period 70 to 100 days after

planting.

Water stress periods in treatments 5 and 6 were imposed by

covering the plots with portable rain shelters (40-70 and 70-110 days,

respectively) only when rain was expected.

'Georgia Jet' and 'Yellow Jewel' sweet potato varieties were

harvested on 4 and 15 Nov 1982, respectively. Potato tuber fresh

weights were obtained from the two central rows of each subplot by
2
taking an area of 4.0 m2. The tubers were graded according to commer-

cial specifications, and subsamples were taken directly to a forced-

draft dryer for several days and then weighed to obtain potato dry






29


2
matter yields. Above-ground vegetation was measured from 1.70 m
2
of the 4.0 m used for harvestable yield. The whole sample was

dried at 700C.

In both varieties, root length densities were measured at harvest

time in treatment 1 (rainfed) and treatment 2 (well-irrigated). Root

samples were obtained in a single site (about 10 cm from the main stem)

per subplot of replications 1, 2, and 3. The soil-root samples were

taken in 15 cm increments to a depth of 210 cm with a 5 cm aluminum

tube. Roots were separated from the soil as described for the corn

experiment. Root length densities were obtained by using a grid-

intercept method (Tennant, 1975).












RESULTS AND DISCUSSION


Oat Lysimeter Experiment


Rainfall and Irrigation


Seasonal rainfall distribution and irrigation schedules for the

growing season of 'Florida 501' oat are shown in Fig. 1 and Table 1,

respectively. Rainfall was regularly distributed over the season,

however, five dry periods with durations of 10 to 24 days were ob-

served. Two heavy rainfalls on days 55 and 88 (8.0 and 6.7 cm) caused

water losses by drainage in all treatments. Irrigation was used only

between days 75 and 97 (3 through 25 February).


Evapotranspiration


Periodic and seasonal evapotranspiration rates were obtained

from water balance computations. Because periodic actual evapotrans-

pirations for all treatments were nearly the same during the period

0-75 days, average ETa values were used. Figure 2 shows a comparison

between ETa rates for treatments 1 and 3 (rainfed and well-irrigated)

and the potential evapotranspiration rate (ETp) as estimated by Penmans's

(1948) method. Irrigation in treatment 3 increased the ETa rates over

the ETp rates furing the irrigated period (75 to 97 days). In contrats,

the ETa rates of the rainfed treatment almost reached the ETp rate

during the same period of time. Note that the ETa for both treatments

were lower than the ETp rates from the period of 0 to 60 days, During



30






31






3
PLANTING ,NOV. 2 0,1981 HARVEST, MARCH 1,
2. 82


TREATMENT 5
0

0 2.

z2 TREATMENT 4
0

0







(5
O TREATMENT 3
-- 0-____________H P t



TREATMENT 2


7- TREATMENT 1

6-
E
5-

-j4



2-




-10 0 20 40 60 80 100 110
DAYS FROM PLANTING





Fig. 1. Rainfall distribution and irrigation schedules for
'Florida 501' oat dry matter production.






32






Table 1. Irrigation schedule, 'Florida 501' oat, 1981-1982.


Irrigation amount on treatment number
Date Day 2 3 4 5

cm
Feb. 3 75 1.3 1.5

5 77 1.5 -

6 78 0.5 --

7 79 -2.5 1.3

11 80 1.0 -

13 85 1.5 -

15 87 1.0 2.5

20 92 1.5 -

21 93 1.0 2.0

24 96 1.0 -

25 97 1.5 -


TOTAL 5.8 6.0 4.5 5.3









7

6- IRRIGATED (Treat.3) -
RAINFED (Treat. )
ETp
T 5-





IIf
o ,











0 10 20 io 4o 5o 6o 7o 8o 9o 1b0 O
DAYS FROM PLANTING





Fig. 2. Measured average daily ETa for two treatments as compared with daily ETp,
'Florida 501' oat, 1981-1982.
E; *, ,







0 I
130 6o "to 8o 9o 1oo 11

DAYS FROM PLANTING





Fifi. 2. Measured average daily ETa for two treatments as compared with daily ETp,
'Florida 501' oat, 1981-1982.

ii>





34



the early part of the season, incomplete canopy cover could not

totally intercept solar radiation, so ETa rates were lower than the

potential rates. When the crop develops a leaf area index of 4 almost

90% of the solar radiation is intercepted (Ritchie, 1972) and ETa

rates reach or exceed ETp expecially when there is advection of

energy from non-irrigated areas.

The water-use coefficients (Kc) for oat dry matter production

were calculated assuming that oat plants in the early part of the

season were growing with no severe stress. Experimental ETa data of

well irrigated treatment 3 and estimated ETp values were related in

the following expression:


Kc ETa (-) (16)
ETp

Figure 3 shows the variation of Kc coefficients over the growing

season. Crop coefficients eventually exceeded 1 at the end of the

growing period. At harvest, the oat plants had developed a full

canopy, and were in transition from vegetative to reproductive stages.

Advection of energy from unirrigated areas surrounding the lysimeter

could have accounted for the increase in ETa over ETp.


Dry Matter Yield Response


Irrigation amount, seasonal evapotranspiration and oat dry

matter yields for all treatments are summarized in Table 2. All

irrigated treatments produced higher yields (0.01 level) than the

rainfed treatment. The highest dry matter yield was obtained in

irrigated treatment 3.














1.2-



1.0O


0.8-



- 0.6-
U-
LL
W
0
S0.4-
QL
0

0.2-



0
10 26 30 40 5b 6'0 0 e'o 6o 1oo
DAYS AFTER PLANTING


Fig. 3. Crop water use coefficients for oat dry matter production.
Z ,^^






36






Table 2. Effect of water management on dry matter yield of 'Florida
501' oat, 1981-1982.



Treatmentt Irrigation ETa Dry matter yield
-i
cm --- g ha-
1 0 20.50 1450 d

2 5.80 (6)$ 25.70 4300 ab

3 6.00 (4) 26.00 4520 a

4 4.50 (2) 24.60 3610 c

5 5.30 (3) 25.00 4050 b

tSee p.23 for description of treatments.
SNumbers in parenthesis are the number of irrigation applications.
Means followed by the same letter are not significantly different
(0.05 level, Duncan's Multiple Range Test).






37


The relationship between dry matter yield and seasonal evapo-

transpiration as well as irrigation was determined by simple linear

regression analysis. The data were plotted (Fig. 4) as suggested by

Stewart and Hagan (1973). Dry matter yield was linearly related to

ETa and irrigation. Note that the two linear relationships are not

superimposed since a given irrigation amount did not produce an equal

increase in evapotranspiration.

The dry matter-ETa relationship conforms with the findings of

several researchers (Arkley, 1963; Hanks et al., 1969; Hillel and

Guron, 1973; Stewart et al., 1977; Fischer, 1979). It is more common

to find a curvilinear (concave downward) relationship between yield

and irrigation (Yaron, 1971; Shipley and Regier, 1975, 1976). Stegman

et al. (1980) indicated that the irrigation function is typically

curvilinear and curves away from the ET function as irrigation amounts

increase.

The yield-irrigation linear relationship obtained in the present

research can be explained in terms of water management factors. The

irrigation rates in each irrigated treatment were not high enough to

produce large drainage, i.e., drainage losses were proportional to

irrigation amounts. In addition, nutrient losses or root aeration

problems were not observed.

The regression coefficient of both dry matter-ET and dry matter-

irrigation relationships are called crop-water production functions.

The irrigation production function is less than or equal to the ET

production function. The goal in water management is to obtain an

irrigation function as near the ET function as possible.















/'o
b "

4000- -







)/ -- Y = 1427 + 499 (IRRIG)
J/ (R2 = 0.99)
2000
'2000- --- Y = -9966 +556 (ET)
w (R2 = 0.97)

S 2 3 4 5 6 7 8 9
>1000- IRRIGATION DEPTH (cm)
0
I--
0


18 20 22 24 26 28 30
EVAPOTRANSPIRATION (cm)



Fig. 4. Dry matter yield of 'Florida 501' oat versus seasonl ET and seasonal
irrigation (IRRIG).
Co





39



The ET and irrigation functions obtained in this research were
-1 -1
556 and 499 kg ha cm respectively. The ratio of these two values

(499/556 = 0.90) provides an estimate of the fraction of the water

applied that was used by the crop as ET (Hammond et al., 1981a).

Thus, the smaller the slope of the irrigation function the less effi-

cient was the water management strategy during the season. The rather

large seasonal water-use efficiency of 90% could be attributed to a

good combination of irrigation scheduling and rainfall.

Figure 5 shows the dry matter and ET data plotted as yield reduc-

tion in response to decreasing seasonal ET. The slope of this rela-

tionship (Ky) called the yield response factor (Doorenbos and Kassam,

1979) gives the unit of relative dry matter reduction per unit of rela-

tive ET deficit. Thus, from this relationship, it is observed that a

seasonal ET deficit of 20% will produce a 64% reduction in dry matter

production (3.20 x 0.20 = 0.64). A relative ET deficit of 20% repre-

sents an absolute ET deficit of 5.2 cm (0.20 x 26 cm). A relative dry

matter yield decrease of 64% represents an absolute yield decrease of

2893 kg ha-1 (0.64 x 4520 kg ha- ). Dividing the absolute dry matter

decrease by the absolute ET deficit (2893/5.2) gives 556 kg ha- cm,

which is the value of the ET production function previously obtained

(Fig. 4).


Water Balance


Measured periodic water balance values for the oat experiment

are presented in Table 3. Total water balance for the growing season

is summarized in Table 4. Water depletions were calculated from





40











(1- ETa ETm)
LO 0.6 04 02 0









0
/ .
-/ /
/















Ky -.20
/














R = 0.97










Fig. 5. Relationship between oat relative dry matter yield
decrease (1 Ya/Ym) and the relative seasonal ET
deficit ( ETa/ETm)
/ --0.4 (1-Ya/Ym)


/
/
-/ / -0.6
/













Fig. 5. Relationship between oat relative dry matter yield
decrease (1 Ya/Ym) and the relative seasonal ET
deficit (1 ETa/ETm).







41



Table 3. Periodic measured water balance during the vegetative growth
period of 'Florida 501' oat, 1981-1982.



Treatmentt Inputt AS Depletion Drainage ETa

cm
0-6 days (ETp = 1.48 cm)#

1-5 0.254 -0.187 0.441 0.133 0.308
6-32 days (ETp = 5.16 cm)
1-5 3.28 -0.548 3.83 1.08 2.75
32-44 days (ETp = 2.10 cm)
1-5 2.69 +0.77 1.92 0.17 1.75
44-46 days (ETp = 0.39 cm)
1-5 0.0 -0.36 0.36 0.0 0.36
46-57 days (ETp = 1.90 cm)
1-5 10.62 +1.95 8.67 7.31 1.36
57-64 days (ETp = 1.30 cm)
1-5 0.28 -1.61 1.89 0.0 1.89
64-75 days (ETp = 2.58 cm)
1-5 2.82 -1.83 4.65 0.0 4.65
75-95 days (ETp = 5.61 cm)
1 9.88 -1.24 11.12 5.45 5.67
2 14.13 -1.00 15.13 5.84 9.29
3 13.88 -0.74 14.62 5.31 9.31
4 13.88 -1.29 15.17 6.00 9.17
5 14.88 -0.79 15.67 6.17 9.50
95-98 days (ETp = 0.98 cm)
1 0.0 -1.24 1.24 0.0 1.24
2 1.0 -0.82 1.82 0.0 1.82
3 1.50 -1.45 2.95 1.14 1.81
4 0.0 -1.15 1.15 0.0 1.15
5 0.0 -1.25 1.25 0.0 1.25






42




Table 3--continued.


Treatmentt Inputs AS Depletion Drainage ETa

cm
98-101 days (ETp = 0.83 cm)
1 0.0 -0.10 0.10 0.0 0.10
2 0.0 -1.53 1.53 0.0 1.53
3 0.0 -1.82 1.82 0.0 1.82
4 0.0 -1.18 1.18 0.0 1.18
5 0.0 -1.21 1.21 0.0 1.21


tSee p.23 for description of treatments.
tRainfall plus irrigation.
ETa obtained as the difference of measured soil water depletions and
measured drainage.
Periodic water balance averaged for all treatements from the period
of 0-75 days.
#From daily estimates, Penman method. See Appendix Table 27.











Table 4. Total seasonal water balance of 'Florida 501' oat, 1981-1982.


Profile
Treatmentt Total inputt AS water Drainage ETa ETp
depletion
cm
1 29.3 -4.78 34.08 13.58 20.5 22.33

2 35.1 -5.16 40.26 14.56 25.7 22.33

3 35.3 -5.83 41.13 15.13 26.0 22.33

4 33.8 -5.43 39.23 14.63 24.6 22.33

5 34.6 -5.05 39.65 14.65 25.0 22.33


TSee p.23 for description of treatments.
SRainfall plus irrigation.






44



neutron readings taken at periods of time given in Table 3. As

mentioned earlier, because irrigation was used only in the last part

of the season (from day 75 to day 97), an average (over treatments) of

periodic water balance values was computed for the 0-75 days period.

Periodic actual ET values were calculated as the difference be-

tween periodic depletion (Table 3) and periodic drainage. Large

drainage losses (90% of the season total) occurred following two heavy

rainfalls (days 55 and 85). Average per treatment irrigation input was

5.4 cm; average ETa increase was 4.83 cm or 4.83/5.4 = 0.89 efficiency

average drainage increase was 1.16 cm, but some of this came from a

decrease in stored water = average of 0.59 cm. Thus, the average in-

crease in drainage attributed to irrigation was (1.16 0.59) 0.57 cm

or 0.57/5.4 = 0.11. Hence, 89% of irrigation contributed to an increase

in ETa and 11% went to an increase in drainage. The 89% irrigation-use

efficiency compares favorably with the 90% efficiency calculated by

the ratio of the two production functions.


Corn Experiment


Rainfall and Irrigation


The rainfall and irrigation distributions for the growing season

of the 'McCurdy 84aa' corn are shown in Fig. 6 and Table 5. All treat-

ments received an initial irrigation of about 1.27 cm on 4 March which

was recorded as rainfall. The maximum daily rainfall, 9.88 cm, was

recorded on day 42. A rain-free period between 55 and 75 days permitted











4 TREATMENT 3
E
F 2-

TREATMENT 2
2-
0 B Bill R I 1I
TREATMENT I
PLANTING, TREATMENT 1 HARVEST,
8 FEB. 26,1982 JULY 9,1982

6-

3 4-

2lu

0-
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
FEB MARCH I APRIL | MAY JUNE I JULY



Fig. 6. The distribution and amount of rainfall and irrigation during the
growing season of 'McCurdy 84aa' corn, 1982.







46





Table 5. Irrigation schedule, 'McCurdy 54aa' corn, 1982.


Irrigation amount on treatment number
Date Day 2 3

cm
March 18 21 1.02 1.02
April 8 42 1.02 1.02
17 51 1.27 1.27
21 55 1.78 1.78
May 2 66 2.54
6 70 2.54
9 73 1.27
12 76 1.52
13 77 3.05
14 78 1.90 0.38
17 81 2.03 2.03
21 85 2.03 2.03
30 85 2.03 2.03
June 7 102 1.90 1.90
10 105 0.63 0.63
11 106 2.54 2.54

TOTAL 25.51 19.17





47



the development of a planned 2-week water stress period for treat-

ment 3.

Differential irrigation treatments began on day 21 when treat-

ments 2 and 3 were irrigated. Thereafter, irrigations were accomplished

for the two irrigated treatments when the required conditions were met.

The total seasonal amount of irrigation applied in treatments 2 and 3

were 25.5 and 19.2 cm, respectively.


Yield Response


Dry matter yield response to irrigation and estimated ETa fit

linear functions very well (Figs. 7 and 8). In contrast, grain yield

response gave a good fit only under nitrogen sufficient conditions

(Figs. 9 and 10). For dry matter, nitrogen stress reduced irrigation

and ET regression coefficients by 24 and 39%, respectively. The

associated reductions for grain yields were 57 and 46%. The grain ET

function under nitrogen sufficient conditions (457 kg ha- cm ) is in

agreement with calculated values from corn grain data of Hillel and

Guron (1973) in Israel (540, 440, and 450 kg ha- cm-). However,

Hammond et al. (1981b) reported a larger ETa production function

(661 kg ha- cm ).

Although treatment 3 was water stressed during a critical growth

stage, there was no significant departure from linearity in any of the

four functions in Figs. 7 and 9. However, there were only three water

management levels, and the rainfed treatment was a rather severe

drought treatment. Tanner and Sinclair (1983) and Sinclair et al. (1984)

state that the relationships between grain yields and ET are essentially











30


2
0 --- Y = -11799 + 722 (ETa); R = 0.99
25 2
-- Y = 7022 + 566 (IRRIG); R = 0.99




20-




E-I
g 15-



10




0 4 8 12 16 20 24 28
5
SEASONAL IRRIGATION (cm)

L^ -, ---- --- ---- ---- --- ---- -- -- ^- _..-- ___, ---
22 26 30 34 38 42 46 50 54 58

SEASONAL EVAPOTRANSPIRATION (cm)



Fig. 7. 'McCurdy 84aa' corn dry matter yield versus seasonal ET and seasonal irrigation (I',I.C),
under nitrogen sufficient conditions, 1982. (All corn yield data---Fi;s. 7 through 12--
courtesy of Drs. J. M1. Bennett, L. C. Hammond, J. W. Jones, P. S. C. Rao, and '!r. L.
utti) .
















20
S--- Y = -8670 + 552 (ETa); R = 0.99
S-- Y = 5482 + 351 (IRRIG); R2 = 0.99
S 15




S10




P, 5
S4 8 12 16 20 24 28

SSEASONAL IRRIGATION (cm)
0
22 26 30 34 38 42 46 50 54

SEASONAL EVAPOTRANSPIRATION (cm)

Fig. 8. 'McCurdy 84aa' corn dry matter yield versus seasonal ET and seasonal irriga-
tion (IRRIG), under nitrogen stress conditions, 1982.












12 /*




0 /

7 8- -- Y = -8882 + 457 (ETa)
.r / R2 = 0.93
-/ -- 2980+342(IRRIG)
S6- R / 2R= 0.94
3 _J

> 4-

<1 "1 IlIlI i I _
D 4 8 12 16 20 24 28
S2- SEASONAL IRRIGATION (cm)


O- t' I a I I 1
22 26 30 34 38 42 46 50 54
SEASONAL EVAPOTRANSPIRATION (cm)


Fig. 9. 'McCurdy 84aa' corn grain yield versus seasonal ETa and seasonal
irrigation (IRRIG) under nitrogen sufficient conditions, 1982.














CI
0 8.

00

6-

C/ -- Y = -2812 + 248 (ETa); R2 = 0.83
w .. -- Y = 3581 + 148 (IRRIC); R = 0.81

z
) 0 4 8 12 16 20 24 28
S 2- SEASONAL IRRIGATION (cm)




0 22 26 30 34 38 42 46 50 54
SEASONAL EVAPOTRANSPIRATION (cm)


ig. 10. 'McCurdy 84aa' corn grain yield versus seasonal ETa and seasonal irrigation (IRRIG)
under nitrogen stress conditions, 1982.





52




linear due to the fact that crop harvest indexes remain relatively

constant over levels of ET deficits.

Curvilinear (concave downward) relationships between grain yield

and irrigation are commonly found when treatments receive excessive

irrigation (Stegman et al., 1980). Yield reductions at high levels of

irrigation can result from poor soil aeration and leaching of plant

nutrients.

Note that the irrigation production coefficients were lower than

the ET production coefficients (Figs. 7, 8, 9, and 10). The ratio of

these coefficients (irrigation/ET) provides an estimate of the fraction

of the seasonal irrigation which was used to increase evapotranspiration.

Thus, these ratios represent the average irrigation-use efficiency

achieved in the specific experiment. The efficiencies were higher under

nitrogen sufficient conditions (0.78 and 0.75 for dry matter and grain

responses, respectively). The respective efficiencies under nitrogen

stress were 0.62 and 0.60. Hammond (1981b) stated that the goal in

irrigation management is to achieve a ratio equal to one. This ideal

is not achieved very often, because some of the irrigation may remain

stored in the root zone or rainfall may displace it from the root zone.

The data from Figs. 9 and 10 were replotted in Fig. 11 to more

clearly show the interaction of nitrogen and water stress.

The relative corn grain yield reduction as a function of relative

ET deficit is shown in Fig. 12 for N sufficient conditions. The slope

(Ky) of this relationship, called the yield response factor by Doorenbos

and Kassam (1979), indicates the unit of relative yield reduction per

unit of relative ET deficit. Thus, under rainfed conditions (treatment






53






12000- N1/

Y = -8882 + 457 (ET) /
10000- 2/
8R = 0.93
.NO
:8000. ,


S6000 -

z lo
z 4000- ."
SY = -2812 + 248 (ET)
2 2000-, R = 0.83

24 28 32 36 40 44 48 52
SEASONAL EVAPOTRANSPIRATION (cm)



12000
Ni

10000 Y2980+ 342(IRRIG)
WR=0.94
8000- NO
-J
$u o
6000-
z

4000-
z Y=3582+ 148 (IRRG)
S2000 R2 0.81
0

0 4 8 12 16 20 24 28 32
SEASONAL IRRIGATION (cm)


Fig. 11. 'McCurdy 84aa' corn evapotranspiration and irrigation
production function under nitrogen sufficient (N ) and
nitrogen stress (N0) conditions, 1982.









ETo
( ETm )
1.0 0.8 0.6 0.4 0.2 0



0.2


/ 0.4

/ (t._- o)
Ym
/ -0.6



/ Ky 1.72 0Q.8
/ R2=0.93
V _____1.0





Fig. 12. Relative corn grain yield reduction (1 Ya/Ym) versus relative ET deficit
(1 ETa/ETm) under nitrogen sufficient conditions.
Un






55



1), the relative seasonal ET deficit of 43% produced a relative grain

yield reduction of 74% (1.72 x 0.43 = 0.74). The relation of Ky to

the production coefficient is given by the following expression:


ETcoefficient = Ky (Ym/ETm). (17)


Thus, for the corn data in Fig. 6, we have

-1 -1 -1
ET ffiit = 1.72(12300 kg ha /46.35 cm) = 457 kg ha cm- .
coefficient

Root Length Density Distribution


Roots of 'McCurdy 84aa' corn were sampled on day 103 for all

treatments. The samples were taken in four sites for each treatment

of replication IV. Root length density distributions are given in

Fig. 13. There was a significant water management x nitrogen x depth

interaction. No other factors were significant. The rainfed nitrogen

stress treatment (WINO) had the lowest concentration of roots in the

0-15 cm and 30-60 cm layers. There were no differences in root density

distributions in the 60-105 cm layer, but there were differences in

the 105-120 cm layer. No roots were found in the well irrigated treat-

ment ( 2), and lower root density values were found with nitrogen stress

than with sufficient nitrogen in both the rainfed and limited irrigation

(3) treatments. Apparently, the stunted growth resulting from a combi-

nation of water and nitrogen stress reduced photosynthate production and

delivery to the root system. In addition, root penetration of the soil

profile could have been restricted due to increasing soil strength with

drying. Persad (1982) found that root distribution of corn plants, in

the same area in 1981, was restricted by the presence of a tillage pan

at the 30-50 cm depth.





56











ROOT LENGTH (cm cri of soil)


I
E 0


J 6 TREAT. I TREAT.3 TREAT. 2


.. 120 NO NO NO



I2 60 TREAT I -TREAT.3 TREAT.2


S12 NI NI NI










Fig. 13. Root length density distribution of 'McCurdy 84aa' corn
on day 103 after planting, 1982.






57



Water Balance


Periodic water balance data showing a comparison between measured

and simulated profile water depletions for 'McCurdy 84aa' corn are

presented in Table 6. The seasonal water balances calculated by the

summation of the measured or estimated periodic water balance data are

summarized in Table 7. Measured periodic water depletions in Table 6

were calculated from neutron readings. Estimated ETa,drainage, and

water depletion were obtained by simulation. As shown in Table 6,

large drainage losses occurred in the first 43 days of the growing

season (63, 55, and 58% of the total estimated drainage in treatments

1, 2, and 3, respectively). In general, there was close agreement

between measured and estimated periodic depletions for both nitrogen

conditions. However, lack of agreement was particularly observed in

periods 75-81 and 81-84 days for the well irrigated treatment 2 under

nitrogen sufficient conditions. Measured water depletion for the

first period (75-81 days) was 0.42 cm while the estimated depletion

was 3.55 cm. This indicates that for measured depletion most of the

water input was stored in the soil, and that the combined ETa and drainage

was only 0.42 cm for the 6-day period. This, of course, is not reason-

able because the simulated ET for the same period was 3.43 cm. In

contrast, the measured and estimated depletions for the next period

(81-84 days) were 5.96 and 1.91 cm, respectively. For the measured

case it would have to be a considerable drainage event since the

estimated ET value was only 1.69 cm. Thus, the change from low to

high depletion is not reasonable and perturbations in the measured






58




Table 6. Periodic water balance during the growth period of 'McCurdy
84aa' corn, 1982.


Measured water Estimated water
Treatmentt Inputt AS Depletion ETa Drainage AS Depletion
cm
0-19 days, N0 and N! (ETp = 6.90 cm)


1-3 7.04 0.02 7.02 1.85 5.64 -0.45 7.49
19-40 days, N1 (ETp = 7.78 cm)

1 8.01 0.16 7.85 3.50 4.89 -0.38 8.39
2 9.04 0.11 8.93 4.56 4.96 -0.48 9.52
3 9.04 0.16 8.88 4.53 5.00 -0.49 9.53
19-43 days, NO (ETp = 8.65 cm)

1 23.05 2.78 20.27 3.88 16.82 2.35 20.70
2 25.10 2.72 22.38 5.06 17.69 2.35 22.75
3 25.79 2.81 22.98 4.98 18.44 2.37 23.42
40-43 days, N1 (ETp = 0.87 cm)

1 15.04 2.81 12.23 0.59 11.70 2.75 12.29
2 16.06 2.70 13.36 0.74 12.23 3.09 12.97
3 16.06 2.75 13.31 0.69 12.15 3.22 12.84
43-57 days, NO (ETp = 6.06 cm)

1 4.38 -5.56 9.94 3.11 5.18 -3.91 8.29
2 7.43 -4.10 11.53 4.93 5.12 -2.62 10.05
3 7.43 -4.00 11.43 4.77 5.16 -2.50 9.93
43-57 days, N1 (ETp = 6.60 cm)

1 4.38 -5.96 10.34 3.30 5.04 -3.96 8.34
2 7.43 -3.88 11.31 5.35 5.00 -2.92 10.35
3 7.43 -4.12 11.55 5.15 5.06 -2.78 10.21
57-65 days, N1 (ETp = 3.06 cm)

3 2.76 -1.49 4.25 2.84 0.24 -0.32 3.08





59




Table 6--continued.


Measured water Estimated water
Treatmentt Inputl AS Depletion ETa Drainage AS Depletion
cm
57-68 days, NO (ETp = 4.43 cm)

2 5.30 0.09 5.21 4.00 0.43 0.87 4.43
3 2.76 -1.58 4.34 3.43 0.54 -1.21 3.97
57-68 days, N1 (ETp = 4.43 cm)

2 5.30 0.74 4.56 4.51 0.37 0.42 4.88
57-71 days, NO (ETp = 5.90 cm)

1 2.87 -1.99 4.86 3.21 0.05 -0.39 3.26
57-71 days, N1 (ETp = 5.90 cm)

1 2.87 -1.92 4.79 3.27 0.11 -0.51 3.38
65-68 days, N1 (ETp = 1.37 cm)

3 0 -1.20 1.20 0.88 0.01 -0.89 0.89
68-71 days, NO (ETp = 1.47 cm)

2 2.64 1.30 1.34 1.41 0.54 0.69 1.95
3 0.10 -1.31 1.41 0.69 0.04 -0.63 0.73
68-71 days, N1 (ETp = 1.47 cm)

2 2.64 1.06 1.58 1.61 0.36 0.67 1.97
3 0.10 -0.46 0.56 0.66 0.02 -0.58 0.68
71-73 days, N (ETp = 0.99 cm)

3 0 -0.15 0.15 0.38 0.09 -0.38 0.38
71-73 days, N1 (ETp = 0.99 cm)

1 0 -0.94 0.94 0.30 0 -0.30 0.30
3 0 -0.90 0.90 0.37 0 -0.37 0.37
71-74 days, N0(ETp = 1.53 cm)

2 1.27 -1.87 3.14 1.47 0.29 -0.49 1.76
71-74 days, N1 (ETp = 1.53 cm)

2 1.27 -1.51 2.78 1.68 1.04 -1.45 2.72






60



Table 6--continued.


Measured water Estimated water
Treatmentt Inputt AS Depletion ETa Drainage AS Depletion
cm
71-76 days, NO (ETp = 2.53 cm)

1 0 -1.49 1.49 0.72 0.02 -0.74 0.74
73-74 days, N1 (ETp = 0.54 cm)

3 0 -0.20 0.20 0.31 0 -0.31 0.31
73-76 days, NO(ETp = 1.54 cm)

3 3.05 2.02 1.03 0.78 0.20 2.07 0.98
73-76 days, N1 (ETp = 1.54 cm)

1 0 -0.32 0.32 0.38 0.01 -0.39 0.39
74-75 days, N1 (ETp = 0.53 cm)

2 1.52 -1.31 2.83 0.50 0.09 0.93 0.59
74-76 days, N1 (ETp = 1.00 cm)

3 3.05 2.36 0.69 0.48 0.35 2.22 0.83
74-81 days, NO(ETp = 3.55 cm)

2 5.46 2.18 3.28 3.37 0.58 1.51 3.95
75-81 days, N1(ETp = 3.05 cm)

2 3.94 3.52 0.42 3.43 0.12 0.39 3.55
76-81 days, N (ETp = 2.55 cm)

1 0 -0.71 0.71 0.51 0 -0.50 0.50
3 2.41 0.90 1.51 2.23 0.03 0.15 2.26
76-81 days, N1 (ETp = 2.55 cm)

1 0 -0.73 0.73 0.47 0 -0.47 0.47
3 2.11 0.03 2.08 2.31 0.01 -0.21 2.32
81-82 days, N1 (ETp = 0.47 cm)

3 0 -0.89 0.89 0.43 0 -0.43 0.43






61



Table 6--continued.


Measured water Estimated water
Treatmentt Inputs AS Depletion ETa Drainage AS Depletion

cm
81-84 days, NO(ETp = 1.47 cm)

2 2.03 -2.66 4.69 1.38 0.76 -0.11 2.14
3 2.03 -0.60 2.63 1.31 0.03 0.69 1.34
81-84 days, N (ETp = 1.47 cm)

1 0 -0.01 0.01 0.25 0 -0.25 0.25
2 2.03 -3.93 5.96 1.69 0.22 0.12 1.91
81-88 days, N0(ETp = 3.28 cm)

1 6.32 2.96 3.36 1.25 0.13 4.94 1.38
82-84 days, N1 (ETp = 0.99 cm)

3 2.03 -0.56 2.59 0.89 0.25 0.89 1.14
84-88 days, NO (ETp = 1.81 cm)

2 6.31 3.73 2.58 1.70 2.64 1.97 4.34
3 6.32 3.88 2.44 1.66 0.22 4.17 2.15
84-88 days, N1 (ETp = 1.81 cm)

1 6.32 3.27 3.05 1.01 0 5.19 1.13
2 6.31 4.42 1.89 2.08 1.44 1.97 4.34
3 6.32 4.71 1.61 1.73 0 4.42 1.90
88-91 days, NO(ETp = 1.26 cm)

2 0.53 -1.77 2.30 1.19 1.65 -2.28 2.81
3 0.52 -2.69 3.22 1.17 1.67 -2.13 2.66
88-81 days, N1 (ETp = 1.26 cm)

1 0.53 -0.56 1.03 0.88 0 -0.49 1.02
2 0.53 -1.99 2.52 1.45 1.55 -2.44 2.97
3 0.53 -0.94 1.47 1.22 1.56 -2.13 2.66






62



Table 6--continued.


Measured water Estimated water
Treatmenti Inputs AS Depletion ETa Drainage AS Depletion
cm
88-98 days, NO (ETp = 4.19 cm)

1 5.18 1.37 3.81 2.62 1.46 1.10 4.08
91-98 days, NO (ETp = 2.93 cm)

2 6.17 0.32 5.85 2.58 3.88 -0.27 6.34
91-98 days, N1 (ETp = 2.93 cm)

1 4.65 1.63 3.02 1.76 1.33 1.25 3.40
2 6.17 0.08 6.09 2.93 3.40 -0.13 6.30
3 6.17 0.98 5.19 2.57 3.83 -0.23 6.40
91-111 days, NO (ETp = 9.98 cm)

3 17.13 0.32 16.81 8.15 9.94 -0.87 18.10
98-102 days, N1 (ETp = 2.20 cm)

1 0 -0.41 0.41 1.32 0.87 -1.70 1.70
98-105 days, N1 (ETp = 3.95 cm)

2 5.08 -2.07 7.15 3.89 0.58 0.32 4.76
98-111 days, NO (ETp = 7.05 cm)

1 5.89 1.90 3.99 4.00 2.49 -0.61 6.50
2 10.97 2.06 8.91 5.90 5.76 -0.71 11.68
98-111 days, N1 (ETp = 7.05 cm)

3 10.96 -0.38 11.34 5.91 5.75 -0.71 11.67
102-111 days, N1 (ETp = 4.84 cm)

1 5.89 2.38 3.51 2.68 1.60 1.55 4.34
105-111 days, N1 (ETp = 3.10 cm)

2 5.89 0.46 5.43 2.79 4.44 -0.83 6.72
111-124 days, NO (ETp = 5.73 cm)

1 14.60 1.23 13.37 2.77 11.39 0.45 14.15
2 14.61 1.97 12.64 4.26 10.03 0.33 14.28
3 14.61 0.44 14.17 4.21 10.11 0.31 14.30







63



Table 6--continued.


Measured water Estimated water
Treatmentt Inputs AS Depletion ETa Drainage AS Depletion
cm
111-124 days, N1 (ETp = 5.73 cm)

1 14.60 0.66 13.94 2.97 11.29 0.38 14.22
2 14.61 2.18 12.53 4.83 9.55 0.22 14.39
3 14.61 0.69 13.92 4.26 10.03 0.30 14.31
124-127 days, NO (ETp = 1.44 cm)

2 0 -1.18 1.18 1.01 0.49 -1.47 1.47
3 0 -2.06 2.06 1.01 0.49 -1.48 1.48
124-127 days, N1 (ETp = 1.44 cm)

1 0 -1.38 1.38 0.72 0.78 -1.49 1.49
2 0 -1.05 1.05 1.15 0.32 -0.95 0.95
3 0 -0.86 0.86 1.01 0.50 -1.49 1.49
124-130 days, NO (ETp = 3.02 cm)

1 0.41 -1.35 1.76 1.33 1.18 -2.08 2.49
127-130 days, NO (ETp 1.58 cm)

2 0.41 -1.91 2.32 1.11 0.03 -0.73 1.14
3 0.41 -0.50 0.91 1.11 0.03 -0.73 1.14
127-130 days, N1 (ETp = 1.58 cm)

1 0.41 -0.60 1.01 0.79 0.25 -0.63 1.04
2 0.41 -0.50 0.91 1.26 0.01 -0.87 1.28
3 0.41 -0.41 1.24 1.10 0.02 -0.71 1.12


tSee p.23 for description of treatments.
SRainfall plus irrigation.
Soil profile water content on last day of period minus content on first
day.
Estimated ET plus estimated drainage.










Table 7. Total seasonal water balance of 'McCurdy 84aa' corn, 1982.

Length of Measured water Estimated water
Treatmentt season Total input AS$ Depletion ETa Drainage ASt Depletion Etp
cm
N1

1 130 69.74 -1.84 71.58 26.01 43.60 0.13 69.61 55.85

2 130 95.27 -0.05 95.32 46.35 51.31 -2.39 97.66 55.85

3 130 88.92 -0.79 89.71 39.19 50.48 -0.75 89.67 55.85

N0
1 130 69.74 -3.55 73.29 25.21 47.06 -2.53 72.27 55.85

2 130 95.27 0.90 94.37 41.22 55.35 -1.30 96.57 55.85

3 130 88.92 -3.19 92.11 37.73 52.59 -1.40 90.32 55.85

tSee p.23 for description of treatments.
tMeasured and estimated AS calculated from summation of periodic AS (Table 8) rooting depths.
Measured and estimated depletion calculated from summation of periodic depletions (Table 4) of
rooting depths.
Estimated ET and drainage calculated from summation of period ET and drainage (Table 6).






65



water contents must represent instrument and/or operator errors

rather than underestimation or overestimation of both ETa and drainage.

It is obvious that other factors which should influence the results

in a more consistent manner include spatial variation in irrigation

amounts, temporary water storage as a result of variable clay layer

location, variable length of period between soil water measurements,

and variation in precision of the neutron-water content relationship

with water content in the 0-30 cm soil layer. Additional factors

are the assumptions associated with the simulation model: estimated ETa,

crop coefficients, water storage capacity, and residence time of water

in the root zone.

Differences between measured and estimated seasonal depletion

varied from nil (treatment 1, N1) to a maximum of 2.3 cm (treatment

2, N1). The measured input was subject to error. Irrigation was

spatially variable and the pattern changed with wind velocity and direc-

tion. The measured change in soil storage was based on a limited neutron

sample of the spatially (three dimensions) variable soil water content.

The difference (AS) between measured water contents at the beginning

and end of a period plus the measured water input gave the measured

depletion. Simulated depletion resulted from the model allocation of

measured inputs among ETa, drainage, and change in stored water. The

sum of these ETa and drainage components equaled simulation depletion.

Thus, with the same input base, depletion differences (measured and

simulated) must equal differences in theassociated AS values. It follows

that the basic parameter for comparison in these measurements and simula-

tions is the soil water content or depth in the root zone.






66



Figure 14 shows the measured and simulated root zone water con-

tent distributions with time for all treatments under nitrogen suffi-

cient conditions. Simulated water contents were lower than measured

values for all treatments during the first 60 days of the growing

season. From day 60 to the end of the season close agreement was

obtained in irrigated treatments 2 and 3. The data in Fig. 14 suggest

that the model alone could be used successfully in scheduling irriga-

tion. Measured and simulated water content distributions for all

treatments under nitrogen stress conditions are presented in Appendix

Table 28. The pattern of measured and simulated values under nitrogen

stress was similar to that shown in Fig. 14 for nitrogen sufficient

conditions.

The periodic and seasonal changes in soil water content (AS)

are of interest, too. The seasonal values contain a built-up initial

water content profile consisting of soil depth segments added as root

depth increases to maximum. It follows that measured and simulated

profiles will be different since they contain the differences in

measured and estimated water contents in each of these added soil

layers. The dynamics of the system make each layer unique in terms

of the water status at a particular instant. Water contents predicted

by the model and those measured by the neutron meter may agree better

under some circumstances than others. In view of all the factors

involved, the present limited test of validity of the model is very

encouraging.

The actual evapotranspiration (ETa) component of estimated

periodic depletions for treatments 1 (rainfed) and 3 (stressed) were






67









2 FC
TREATMENT lFJNi

S SIMULATED ---- 0 %
W0 0
S8-- MEASURED ooo o


4 1 o 0 15
E o bar
I o
J 0

c. 0--
a 0 20 40 60 80 100 120

E 12 TREATMENT 2 NI1 FC

SIMULATED- / o 0.---o
S8g MEASURED oco oo '


4 o 15
^ obor
bar



0 0 20 40 60 80 100 120
E
1l2- -__A FC
S TREATMENT 31N1 /N -- --

SIMULATED --- / \e
S8 MEASURED ooo
__ / o oi

4- 0 --o
Eda 'o jo-bar
0

0 20 40 0 80 100 120
DAYS AFTER PLANTING





Fig. 14. Measured and simulated root zone water content distri-
bution with time for three water management treatments,
under nitrogen sufficient conditions. 'McCurdy 84aa' corn,
1982.







68

lower than the periodic ETp (Penman, 1948) under both nitrogen condi-

tions throughout the growing season (Table 6). However, the estimated

ETa for the well irrigated treatment 2, under nitrogen sufficient con-

ditions equaled or exceeded the ETp rates for period 57-68 to 91-98

days. This resulted from the selection of crop coefficients (Appendix

Table 23) equal to or greater than 1 during mid-season. Thus, we

should find ETa values greater than ETp during those periods. Even

though estimates of ETa were higher than ETp rates for some periods,

the seasonal ETa estimates were lower than the seasonal ETp for all

treatments under both nitrogen conditions (Table 7). Seasonal ETa

estimates, under nitrogen sufficient conditions were 47, 83, and 70%

of the seasonal ETp for treatments 1, 2, and 3, respectively. Percen-

tages of 45, 74, and 68 were estimated for the same water management

treatments under nitrogen stress conditions.

Water-use efficiency calculations from seasonal water balance data

are given in Table 8. Water-use efficiencies represent the percentage

of seasonal irrigation amounts that were used to increase ETa. As

observed in Table 8, low ETa increases from irrigation were obtained

for both treatment 2 and 3 under nitrogen stress conditions. This re-

sulted in a low water-use efficiency in relation to that obtained under

nitrogen sufficient conditions. Thus, the well-irrigated treatment 2

under nitrogen sufficient conditions gave the highest water-use effi-

ciency (80%). Water stress in treatment 3 for the same nitrogen suf-

ficient conditions resulted in a 14% reduction in water-use efficiency.

In contrast, nitrogen stress in treatments 2 (well-irrigated) and 3

(water-stressed) resulted in similar water-use efficiencies. The above

water-use efficiencies are in close agreement with those previously

estimated by the ratio of water production functions.














Table 8. 'McCurdy 84aa' corn irrigation-use efficiency, 1982.


ETa increase Water-use efficiencyt
Treatment Irrigation NO N1 Average NO N1 Average
cm %
2 25.5 16.0 20.3 18.2 63 80 71

3 19.2 12.5 13.2 12.8 65 69 67


Average 23.3 14.3 16.8 15.5 64 75 69


tWater-use efficiency is defined in this study as the percentage of seasonal irrigation amount
which was used to increase ETa.







70


Peanut Experiment


Rainfall and Irrigation


The rainfall and the detailed irrigation schedule for the growing

season of the peanut experiment are presented in Fig. 15 and Table 9,

respectively. Total rainfall for the growing season differed between

irrigated treatments 2 and 3 because of the sheltering study of treat-

ment 3 (dry cycle) on genotypes UF 781181-1217, PI 383426, Early

Bunch, and Florunner. A total of 19.14 cm of rainfall was sheltered

from treatment 3 during the sheltering study (from 3 July to 2 August).

To establish the crop, all treatments received three small irrigations

that totaled 2.69 cm and were recorded as rainfall.

Differential irrigation treatments began on day 37 when treatments

2 and 3 were irrigated. Thereafter, irrigation was accomplished for

the two irrigated treatments when the required conditions were met.

Total amounts of irrigation in these two treatments were similar (Table

9). Rainfall was well distributed throughout the season and rainfall

events greater than 2.5 cm were recorded on days 18, 27, 39, 45, 49,

52, 70, 76, 110, and 127. The total amount of rainfall during the

season including the initial irrigation of 2.69 cm was 78.83 cm.


Yield Response


Because the sheltering study in irrigation treatment 3 involved

only four peanut genotypes, two statistical analyses were performed

on the pod yields. The first one considered treatment 1 (rainfed)

and treatment 2 (optimal irrigation) having the 10 peanut genotypes















TREATMENT 3
o DRY CYCLE
z
O WATER STRESS

TREATMENT 2


0.
TREATMENT 1

6 PLANTING, MAY 5,1982 HARVEST, SEPT.20,
1982
E 4





0-
0..,J




0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
MAY JUNE JULY AUGUST SEPT.




Fig. 15. The distribution and amount of rainfall and irrigation during the growing season of peanuts,
1982. -J








72





Table 9. Irrigation schedule, peanuts, 1982.


Irrigation amount on treatment number
Date Day 2 3
cm

June 10 37 1.90 1.90

July 2 59 1.78 1.78

July 9 66 1.68

Aug. 10 98 1.52 2.54

Sept. 2 121 2.03 2.03

TOTAL 8.91 8.25








73



as subtreatments. The second analysis considered only the four peanut

genotypes which were in all three water management treatments.

Pod yield (averages of four replicates) for all main plot treat-

ments are summarized in Table 10. As there was no significant inter-

action between irrigation and genotypes, average yields across geno-

types can be used in analysis of irrigation and ETa production func-

tions. The irrigation amount of 8.91 cm did not increase pod yield

significantly. These results indicate that seasonal rainfall distri-

bution adequately supplied the peanut evapotranspirative needs. The

results are in agreement with the findings of Varnell et al. (1976) in

which supplying 10 cm of water in addition to the 70 cm of natural

rainfall did not improve the yield of Florunner.

As indicated in the footnotes of Table 10, the irrigation regres-
-1 -1
sion coefficients were 55 and 57 kg ha cm for 10 and 4 genotype

averages, respectively, using treatments 1 and 2. The respective ETa
-1 -1
regression coefficients were 159 and 166 kg ha cm Irrigation-use

efficiencies were about 34% in both cases. The addition of treatment

3 to the 4 genotype averages provided for a linear regression analysis.

The linear function was:


Pod yield = 146 + 99 ETa R2 = 0.19. (18)

All 48 data points comprising each treatment average were used in the

regression analysis. The low coefficient of determination reveals the

high variation in the yield data.

It is interesting that the ETa regression coefficients, 159 and
-1 -1
166 kg ha cm compare favorably with the 162 value obtained by

Hammond et al. (1981b) in 1977 with an irrigation-use efficiency of







74





Table 10. Peanut pod yields and simulated evapotranspiration as
affected by water management, 1982.



Pod yields
Treatment Irrigt ETa Ten genotypest Four genotypest
-I
cm kg ha-
1 45.49 4866 4534

2 8.91 48.55 5353 5043

3 8.25 37.54 3888


tSeasonal rainfall 78.73 cm for treatments 1 and 2 and 60.24 cm
for treatment 3.
VFrom treatments 1 and 2, AY/Irrig = 55 and 57 kg ha cm for 10
and 4 genotype averages, respectively. Respective AY/ETa values
159 and 166 kg ha-1 cm-1.
Pod yield data courtesy of Drs. K. J. Boote, J. M. Bennett, and
L. C. Hammond and Mrs. T. D. Rodriguez.







75




77%. Moreover, under well-irrigated conditions, the estimated ETa

values and pod yields in 1977 were nearly the same as in 1980. Clearly,

more irrigation water was applied in the present study than needed during

the relatively wet growing season of 1982. Theoretically, the amount

of irrigation needed in 1982, assuming a 77% efficiency would be 4 cm.
-1 -1
An ETa regression coefficient of 120 kg ha cm was calculated

from the data of Pallas et al. (1979). Their study was conducted on

Florunner peanuts at Tifton, Georgia.

The Ky values of Doorenbos and Kassam (1979) for the 10 and 4

genotype averages were 1.44 and 1.60, respectively. Thus, a 20% ETa

deficit would predict respective yield reductions of 29 and 32%.


Water Balance


Periodic and seasonal water balance data are presented in Tables

11 and 12. Seasonal balances were the summation of periodic water

balance values. Measured water depletions were calculated from rainfall

and irrigation inputs and neutron readings taken in the Florunner

peanut in two replications of each water management treatment. Estimated

depletion and its components, ETa and drainage, were obtained by simula-

tion.

The data of Table 11 show that there was a close agreement between

measured periodic and simulated depletion values throughout the growing

season. The largest difference was observed on treatment 3 during the

induced stress period (60-83 days) where the simulated depletion was

1 cm lower than measured. After the end of stress, the profile remained

at a water content level below the other treatments for about 9 days







76



Table 11. Periodic water balance during the growth period of ten
peanut genotypes, 1982.


Measured water Estimated water
Treatment Inputt ASt Depletion ETa Drainage AS Depletion
cm
0-16 days (ETp = 6.06 cm)
1 2.69 0.37 2.32 2.44 0 0.25 2.44
2 2.69 0.37 2.32 2.44 0 0.25 2.44
3 2.69 0.37 2.32 2.44 0 0.25 2.44
16-36 days (ETp = 11.31 cm)
1 9.50 -0.26 9.76 4.88 5.39 -0.77 10.27
2 9.50 -0.30 9.80 4.93 5.11 -0.54 10.04
3 9.50 -0.30 9.80 4.93 5.11 -0.54 10.04
36-41 days (ETp = 2.61 cm)
1 5.08 1.50 3.58 1.23 2.26 1.59 3.49
2 6.98 1.53 5.45 1.92 3.41 1.65 5.33
3 6.98 1.53 5.45 1.92 3.41 1.65 5.33
41-48 days (ETp = 3.22 cm)
1 8.15 -0.83 8.98 2.73 5.84 -0.42 8.57
2 8.15 -0.37 8.52 2.74 5.79 -0.38 8.53
3 8.15 -0.37 8.52 2.74 5.79 -0.38 8.53
48-55 days (ETp = 3.00 cm)
1 7.13 0.51 6.62 2.64 4.04 0.45 6.68
2 7.13 0.49 6.64 2.64 4.01 0.45 6.68
3 7.13 0.49 6.64 2.64 4.04 0.45 6.68
55-60 days (ETp = 2.57 cm)
3 1.90 -1.25 3.15 2.52 0.40 -1.02 2.92
55-70 days (ETp = 7.01 cm)
1 2.21 -3.99 6.20 5.09 0.89 -3.77 5.98
2 5.66 -1.91 7.57 6.83 1.06 -2.23 7.89
60-83 days (ETp = 9.60 cm)
3 0 -5.32 5.32 4.29 0.04 -4.33 4.33






77




Table 11--continued.


Measured water Estimated water
Treatment Inputt ASS Depletion ETa Drainage AS Depletion

cm
70-73 days (ETp = 1.35 cm)
1 5.69 4.66 1.03 1.17 0.12 4.40 1.29
2 5.69 3.33 2.36 1.41 0.65 3.63 2.06
73-77 days (ETp = 1.38 cm)
1 5.33 1.58 3.75 1.51 2.22 1.60 3.73
2 5.33 0.38 4.95 1.51 2.92 0;90 4.43
77-79 days (ETp = 0.74 cm)
1 0.17 -1.56 1.73 0.81 1.00 -1.64 1.81
2 0.17 -1.56 1.73 0.81 0.85 -1.49 1.66
79-83 days (ETp = 1.70 cm)
1 2.08 -0.55 2.63 1.87 0.78 -0.57 2.65
2 2.08 -1.08 3.16 1.87 0.78 -0.57 2.65
83-90 days (ETp = 2.55 cm)
3 0 -0.61 0.61 0.61 0 -0.61 0.61
83-94 days (ETp = 4.25 cm)
1 5.51 -0.14 5.65 4.67 1.01 -0.17 5.68
2 5.51 -0.18 5.69 4.67 1.00 -0.16 5.67
90-97 days (ETp = 3.05 cm)
3 2.55 0.98 1.57 1.61 0.04 0.90 1.65
94-100 days (ETp = 2.68 cm)
1 0.18 +2.99 3.17 2.95 0.28 -3.05 3.23
2 1.70 -1.20 2.90 2.95 0.02 -1.27 2.97
97-99 days (ETp = 0.90 cm)
3 2.54 1.44 1.10 0.76 0.36 1.42 1.12
99-120 days (ETp = 8.37 cm)
3 11.81 1.59 10.22 7.62 2.76 1.43 10.38
100-106 days (ETp = 2.44 cm)
1 1.85 -0.07 1.92 2.35 0 -0.50 2.35
2 1.85 -0.61 2.46 2.64 0.01 -0.80 2.65







78



Table 11--continued.


Measured water Estimated water
Treatment Inputt ASS Depletion ETa Drainage AS Depletion
cm
106-108 days (ETp = 0.67 cm)
1 2.03 1.35 0.68 0.73 0.10 1.20 0.83
2 2.03 1.51 0.52 0.74 0.01 1.28 0.75
108-116 days (ETp = 3.14 cm)
1 7.57 0.40 7.17 3.29 3.70 0.58 6.99
2 7.57 -0.22 7.79 3.29 4.24 0.04 7.53
116-131 days (ETp = 5.05 cm)
1 10.82 2.12 8.70 4.89 3.86 2.07 8.75
2 12.85 2.06 10.79 4.91 5.92 2.02 10.83
120-139 days (ETp = 5.85 cm)
3 15.24 2.38 12.86 5.46 7.44 2.34 12.90
131-139 days (ETp = 2.50 cm)
1 2.74 -0.88 3.62 2.25 1.47 -0.98 3.72
2 2.74 -1.27 4.01 2.25 1.49 -1.00 3.74


tRainfall plus irrigation.
tSoil profile water content on last day of period minus content on
first day.
Estimated ET plus estimated drainage.












Table 12. Total seasonal water balance, peanut genotypes, 1982.

Length of Measured water Estimated water
Treatmentt season Total inputs AS Depletion ETa Drain AS Depletion ETp
days cm

1 139 78.73 1.22 77.51 45.49 32.97 0.27 78.46 59.09

2 139 87.63 0.97 86.66 48.55 37.30 1.78 85.85 59.09

3 139 68.49 0.93 67.56 37.54 29.42 1.53 66.96 59.09


tSee p.24 for description of treatments.
fRainfall plus irrigation.







80



(90-99 days). In general, there was a consistent equality of neutron-

based and simulated AS values which contrasted with the periodic diver-

gent data obtained in the corn experiment because of a malfunctioning

neutron-meter. For the peanut experiment, the seasonal difference

between measured and simulated depletion was less than 1 cm (Table 12).

Figure 16 also shows a close agreement between measured and simu-

lated root zone water contents for all treatments. Note that drought

from rain-sheltering treatment 3, was severe enough to reduce the root

zone water content to less than 15 bars content. Figure 16 also

reflects the good rainfall distribution during the growing season

(treatment 1). Thus, there was no marked difference in profile water

depth between the rainfed and well-irrigated treatments. Note from

Table 12 that 8.9 cm of irrigation on treatment 2 increased ETa by

3.06 cm. So a low water-use efficiency of 34% (3.06/8.9 = 0.34) was

obtained for the wet peanut growing season.

Simulated ETa values were lower than ETp (Penman, 1948) from

planting to day 55 for all treatments. During mid-season (73-116

days) simulated ETa exceeded ETp in treatments 1 and 2. For the

stressed treatment, it is interesting to note that in the 55-60 days

period preceding the induced water stress, simulated ETa had almost

reached estimated ETp.

During the induced stress simulated ETa was reduced significantly.

After the relief of stress (120-139 days) ETa values were increased

to almost equal ETp. Seasonal ETa estimates were about 77, 82, and

64% of the seasonal ETp for treatments 1, 2, and 3, respectively.








81











S16 TREATMENTI
I4 SIMULATED---- ,
o', FC
2 MEASURED oooo o o,


E 2 (7


0 8
0 0 _______
o 15 bar




S20
TREATMENT 2
SIMULATED ----
t16 b
MEASURED oooo
a. 14 FC


to
20 4I6 80 10 120 4


'-





^ 18. TREATMENT 3
16 1

14' -PERIOD .1' t l FC

Fgo1 MEASURED oooo r/ z w '


14 "- .. ,15 bar


C 20 40 60 80 100 120 140

DAYS AFTER PLANTING





Fig. 16. Measured and simulated root zone water content distri-
butions with time under rainfed (treatment 1), well
irrigated (treatment 2), and irrigated but induced
dry cycle (treatment 3) conditions, peanut genotypes,
1982.
1982.







82


Soybean Experiment


Rainfall and Irrigation


Rainfall and irrigation distribution for the growing season of

'Cobb' soybeans are shown in Fig. 17 and Table 13. All treatments

received three initial irrigations on days 2, 4, and 9 (total = 2.34

cm) and they were recorded as rainfall. There were four drought

periods in which irrigation was used.

Differential irrigation treatments began on day 40 when treatments

4, 5, and 6 were irrigated. Thereafter irrigations were accomplished

for all irrigated treatments when the required conditions were met.

Figure 17 and Table 13 show that irrigations were applied in small

amounts (less than 2.0 cm) to partially refill the depleted rooting

zone. Seasonal rainfall varied with treatment (see Table 13).


Yield Response


The linear regression analysis between soybean yield and seasonal

irrigation is presented in Fig. 18. The rainfed treatment gave unus-

ually high yields when compared to the linear response to irrigation

exhibited by the irrigated treatments. It is the opinion of Drs. K. J.

Boote and L. C. Hammond (personal communication) that the rainfed yields

are biased upward because of yield sampling to avoid the spatial varia-

bility present in the experimental plots. Including the rainfed plots

in the analysis resulted in the following response function:


Y = 2265 + 39 (IRRIG) R2 = 0.59.

















TREATMENT 6
0

TREATMENT 5
0
S 1 TREATMENT 4
0 0
0 1 I | | B TREATMENT 3
0-


0


o ,I
TREATMENT 1
2 6 PLANTING, JULY 1, 1982
HARVEST, NOV. 5, 1982
4






-10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
DAYS AFTER PLANTING
JULY AUGUST SEPTEMBER OCTOBER NOVEMBER




Fig. 17. Distribution and amount of rainfall and irrigation during the growing season of
'Cobb' soybeans, 1982.







84






Table 13. Irrigation schedule, 'Cobb' soybean, 1982.


Irrigation amount on treatment numbers
Date Dayt 2 3 4 5 6

cm

Aug. 9 40 1.52 1.52 1.52

13 44 1.58 1..58 1.58

27 58 1.52 1.52 1.52 1.52

31 62 1.52 1.52 0.84 1.52

Sept.2 64 1.52 1.52 2.03 1.52

16 78 1.78 1.78 1.78 1.78

Oct. 8 100 1.52 1.52 1.78

12 104 1.90 1.90 1.90

18 110 1.27 1.27

20 112 1.78 -

22 114 1.27 1.27

TOTAL 6.34 12.30 10.08 7.49 15.66


tSeason length: 118 days for treatments 1, 2, and 5; 127 days for
treatments 3, 4, and 6. Respective seasonal rainfalls were 58.04
and 54.20 cm.
















3.5
6


3.0- Y = 1838 + 76 (IRRIC); R2 = 0.90
'^0 (EXCLUDING TREATMENT 1) 4


2x 3
I 2.5
() I1
5


S2.0-
H

H

1.5 -




1.0-



0 2 4 6 8 10 12 14 16

SEASONAL IRRIGATION (cm)



Fig. 18. Grain yield response of 'Cobb' soybeans to seasonal amounts of irrigation. (Seed
yield data of Figs. 18 through 21 are courtesy of Drs. K. J. Boote, J. M. Bennett,
and L. C. Hammond).







86


Hammond et al. (1981b) obtained an irrigation response coefficient
-1 -1
of 84 kg ha cm for Bragg and Cobb soybeans in 1978.

Figure 19 shows results of linear regression analyses between

soybean grain yields and simulated seasonal evapotranspiration. Two

linear relationships were well-defined. The first one was obtained

from treatments 6 (well-irrigated), 3, and 2. In these last two

treatments, water stress occurred during vegetative growth. The

second relationship involved treatments 6, 4, and 5. In treatments

4 and 5, water stress occurred during pod set and pod filling stages.

As observed in Fig. 19, the rainfed treatment was not included in the

regression analysis. However, when included with treatments 2, 3, and 6,

the coefficient was decreased to 63 kg ha cm but the determination

coefficient was reduced from 0.91 to 0.84. With all treatments included,

the function was:

Y = 754 + 59 (ETa) R2= 0.65.

The determination coefficients tend to justify the selection of the

two functional relationships in Fig. 19. The existence of two ET produc-

tion functions in the present study is in line with the argument of

Stewart et al. (1976) favoring the potential for a family of curves

depending upon sequencing of ET deficits during the growing season.

Others have emphasized the interaction of water stress and growth

stages (Denmead and Shaw, 1960). Severe yield reduction as a result

of water stress during pod set and pod filling stages of soybeans have

been found by several researchers (Doss et al., 1974; Sionit and Kramer,

1977; Smajstrla and Clark, 1981; Ambak, 1982).













2
3.5- Y = 304 + 75 (ETa); R2 = 0.99

(TREATMENTS 2, 3, AND 6)


3.0 3 6




2.5-

.1::
b5



2
S2..0- Y = -1782 + 132 (ETa); R = 0.99

z (TREATMENTS 4, 5, AND 6)

1.5-




1.0



24 26 28 30 32 34 36 38 40

SEASONAL EVAPOTRANSPIRATION (CM)




Fig. 19. Grain yield response of 'Cobb' soybeans to seasonal ETa.



oo
-^i




Full Text

PAGE 1

CROP GROWTH AND WATER USE IN RELATION TO WATER MANAGEMENT OF WELL-DRAINED SANDS BY DAVID RIESTRA-DIAZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 198A

PAGE 2

To my parents, David and Gabina my wife, Maria Guadalupe, and daughters Nancy, Monica, and Faby, with love and gratitude

PAGE 3

ACKNOWLEDGMENTS The author is indebted to Dr. L. C, Hanunond, chairman of the supervisory committee, for his support, friendship, encouragement, and opportunities given during his training program. He gratefully thanks Drs P. S. C. Rao, A. G. Smajstrla, K. J. Boote, and J. M. Bennett, for their frequent assistance. He is also indebted to Messrs. Alex Clem, Charles Browning, and David Cantlin, for their invaluable technical assistance and friendship. Special thanks are given to Mr. Kamarudin Ambak, former graduate fellow, for his helpful assistance with field work. The author wishes to express his gratitude to Consejo Nacional de Ciencia y Tecnologia (CONACYT) of Mexico for granting the scholarship which enabled him to conduct his studies at the University of Florida. He is indebted to Dr. Enrique Palacios Velez for cooperation and frequent aid given to the author during his studies at the University of Florida. Appreciation is extended to M. C. Gaudencio Diaz Jimenez for his immense contribution to the author's education, and for his introduction of the author to the beauty and challenge of water management in agriculture. Most of all, the author wishes to express his eternal appreciation to his wife, Maria Guadalupe, for all the support, encouragement, understanding, and help provided throughout these long years of hard work, especially during periods of stress and uncertainty. iii

PAGE 4

TABLE OF CONTENTS PAGE ACKNOWLEDGMENTS iii LIST OF TABLES vi LIST OF FIGURES ix ABSTRACT xii INTRODUCTION 1 LITERATURE REVIEW 4 Crop Growth and Water Use 4 Water Management Strategies 9 Water Use Efficiency 14 MATERIALS AND METHODS 18 General 18 Oat Lysimeter Experiment 22 Corn Experiment 23 Peanut Experiment 25 Soybean Experiment 26 Sweet Potato Experiment 27 RESULTS AND DISCUSSION 30 Oat Lysimeter Experiment 30 Corn Experiment 44 Peanut Experiment 70 Soybean Experiment 82 Sweet Potato Experiment 104 iv

PAGE 5

PAGE GENERAL DISCUSSION 124 Crop-Water Production Functions 124 Water Management Strategy 126 Computer-Aided Irrigation Scheduling 127 Future Research Needs 129 CONCLUSIONS 131 APPENDIX 134 LITERATURE CITED 148 BIOGRAPHICAL SKETCH 153 V

PAGE 6

LIST OF TABLES TABLE PAGE 1 Irrigation schedule, 'Florida 501' oat, 1981-1982.. 32 2 Effect of water management on dry matter yield of •Florida 501' oat, 1981-1982 36 3 Periodic measured water balance during the vegetative growth period of 'Florida' 501' oat, 1981-1982 41 4 Total seasonal water balance of 'Florida 501' oat, 1981-1982 43 5 Irrigation schedule, 'McCurdy 84aa' corn, 1982 46 6 Periodic water balance during the growth period of 'McCurdy 84aa' com, 1982 58 7 Total seasonal water balance of 'McCurdy 84aa' corn, 1982 64 8 'McCurdy 84aa' corn irrigation-use efficiency, 1982 69 9 Irrigation schedule, peanuts, 1982 72 10 Peanut pod yields and simulated evapotranspiration as affected by water management, 1982 74 11 Periodic water balance during the growth period of ten peanut genotypes, 1982 76 12 Total seasonal water balance, peanut genotypes, 1982 79 13 Irrigation schedule, 'Cobb' soybeans, 1982 84 14 Periodic water balance during the growth period of nondefoliated 'Cobb' soybeans, 1982 95 15 Total seasonal water balance of nondefoliated 'Cobb' soybeans, 1982 99 16 Irrigation-use efficiency for 'Cobb' soybeans, 1982 103 17 Irrigation schedule, 'Georgia Jet' and 'Yellow Jewel' sweet potato, 1982 106 vi

PAGE 7

TABLE PAGE 18 Yields of 'Georgia Jet' (GJ) and 'Yellow Jewel' (YJ) sweet potato cultivars in response to irrigation treatments, 1982 107 19 Periodic water balance during the growth period of 'Georgia Jet' sweet potato, 1982 116 20 Total seasonal water balance of 'Georgia Jet' sweet potato, 1982 119 21 'Georgia Jet' and 'Yellow Jewel' sweet potato irrigation-use efficiency, 1982 123 22 Comparative "measured" and estimated daily ETa as well as daily ETp, under the criteria of negligible drainage, full crop canopy, no plant water stress, and best subplot management conditions 128 23 Crop coefficients (Kc) used in the com experiment. 128 24 Crop coefficients (Kc) used in the peanut experiment 136 25 Crop coefficients (Kc) used in the soybean experiment 137 26 Crop coefficients (Kc) used in the sweet potato experiment 138 27 Daily potential evapotranspiration rates as estimated by Penman's method, Gainesville, Florida, November 1981-November 1982 139 28 Measured and simulated profile water depths with time for three water management treatments on corn under nitrogen stress conditions 141 29 Measured and simulated profile water depths with time for four water management treatments on 'Cobb' soybeans under nondef oliation conditions 142 30 Measured and simulated profile water depths with time for four water management treatments on 'Georgia Jet' sweet potato 143 31 Soil water parameters, Kendrick fine sand, corn experiment 144 vii

PAGE 8

TABLE PAGE 32 Soil water parameters, Kendrick fine sand, peanut experiment 145 33 Soil water parameters. Lake fine sand, soybean experiment 146 34 Soil water parameters. Lake fine sand, sweet potato experiment 147 viii

PAGE 9

LIST OF FIGURES FIGURE PAGE 1 Rainfall distribution and irrigation schedules for 'Florida 501' oat dry matter production 31 2 Measured average daily ETa, for two treatments as compared with daily ETp, 'Florida 501' oat, 1981-1982 33 3 Crop water-use coefficients for oat dry matter production 35 4 Dry matter yield of 'Florida 501' oat versus seasonal ET and seasonal irrigation (IRRIG) 38 5 Relationship between oat relative dry matter yield decrease (1 Ya/Ym) and the relative seasonal ET deficit (1 ETa/ETm) 40 6 The distribution and amount of rainfall and irrigation during the growing season of 'McCurdy 84aa' corn, 1982 45 7 'McCurdy 84aa' corn dry matter yield versus seasonal ET and seasonal irrigation (IRRIG) under nitrogen sufficient conditions, 1982. (All corn yield data — Figs. 7 through 12 — courtesy of Drs. J. M. Bennett, L. C. Hammond, J. W. Jones, P. S. C. Rao, and Mr. L. Mutti) 48 8 'McCurdy 84aa' corn dry matter yield versus seasonal ET and seasonal irrigation (IRRIG) under nitrogen stress conditions, 1982 49 9 'McCurdy 84aa' corn grain yield versus seasonal ET and seasonal irrigation (IRRIG) under nitrogen sufficient conditions, 1982 50 10 'McCurdy 84aa' corn grain yield versus seasonal ET and seasonal irrigation (IRRIG) under nitrogen stress conditions, 1982 51 11 'McCurdy 84aa' corn evapotranspiration and irrigation production function under nitrogen sufficient (N ) and nitrogen stress (N ) conditions, 1982 7... 53 ix

PAGE 10

FIGURE PAGE 12 Relative corn grain yield reduction (1 Ya/Ym) versus relative ET deficit (1 ETa/ETm) under nitrogen sufficient conditions 54 13 Root length density distribution of 'McCurdy 84aa' corn on day 103 after planting 56 14 Measured and simulated root zone water content distribution with time for the three water management treatments, under nitrogen sufficient conditions. 'McCurdy 84aa' corn, 1982 67 15 The distribution and amount of rainfall and irrigation during the growing season of peanuts, 1982.. 71 16 Measured and simulated root zone water content distributions with time under rainfed (treatment 1) well-irrigated (treatment 2) and irrigated but induced dry cycle (treatment 3) conditions, peanut genotypes, 1982 81 17 Distribution and amount of rainfall and irrigation during the growing season of 'Cobb' soybeans, 1982. 83 18 Grain yield (Y) response of 'Cobb' soybeans to seasonal amounts of irrigation. (Seed yield data of Figs. 18 through 21 are courtesy of Drs. K. J. Boote, J. M. Bennett, and L. C. Hammond) 85 19 Grain yield response of 'Cobb' soybeans to seasonal ETa 87 20 Relative grain yield (Ya/Ym) of 'Cobb' soybeans versus relative seasonal evapotranspiration (ETa/ ETm) 89 21 Relationship between relative grain yield decrease (1 Ya/Ym) and relative seasonal evapotranspiration deficit (1 ETa/ETm) for 'Cobb' soybeans, 1982 91 22 Relationship between relative yield reduction (1 Ya/Ym) and relative seasonal evapotranspiration deficit (1 ETa/ETm) for soybean production at Gainesville, Florida 93 23 Measured and simulated root zone water content distribution with time for treatment 1 (rainfed) and treatment 6 (well-irrigated) nondef oliated 'Cobb' soybeans, 1982 101 X

PAGE 11

FIGURE PAGE 24 The distribution and amount of rainfall and irrigation during the growing season of 'Georgia Jet' and 'Yellow Jewel' sweet potato, 1982 105 25 Yields of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes versus seasonal irrigation (IRRIG) amounts, 1982 110 26 Yields of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes versus seasonal ETa, 1982 Ill 27 Relationship between the realtive sweet potatoes fresh weight yield decrease and the seasonal relative ET deficit for the 'Georgia Jet' variety 113 28 Average root length density distribution of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes at harvest time, 1982 115 29 Measured and simulated root zone water content distribution with time under rainfed (treatment 1) and well-irrigated (treatment 2) conditions, 'Georgia Jet' sweet potatoes, 1982 121 30 Root depth functions used in the simulation model... 134 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CROP GROWTH AND WATER USE IN RELATION TO WATER MANAGEMENT OF WELL-DRAINED SANDS BY DAVID RIESTRA-DIAZ April 1984 Chairman: Dr. L. C. Hammond Major Department: Soil Science Field experiments were conducted to develop water production functions for oats, corn, peanuts, soybeans, and sweet potatoes. Irrigation strategy was to apply small amounts of water to only partially replenish the soil water deficit in the root zone in order to minimize transport of water and nutrients beyond the crop root zone. Simulation model (NITROSIM) was used to estimate evapotranspiration and drainage, and soil water-storage within the root zone. For all crops, yields increased linearly with seasonal irrigation and evapotranspiration. Crop-water production functions for oat dry matter were 556 and 499 kg ha cm""'", respectively, for actual evapotranspiration (ETa) and irrigation. Due to limited data the only production function for peanuts was pod yield versus ETa (162kg ha"""" cm""'"). In the soybean study, there were two ETa production functions which reflected differences in timing of ETa deficits. Irrigating during pod set and pod filling stages resulted in an ETa production function of 75 kg ha "'" cm""'-. Witholding irrigation during these xii

PAGE 13

critical growth stages produced an ETa function of 132 kg ha cm A unit ETa deficit during pod set and pod filling produced about 1.8 times more yield decrease than a unit ETa deficit during vegetative growth. 'Georgia Jet' sweet potato responded to water management while 'Yellow Jewel' did not. The irrigation and ETa production functions for marketable yield of 'Georgia Jet' were 2348 and 3188 kg ha cm respectively. Our water management strategy gave variable results depending on the combination of rainfall events and irrigation inputs. Wateruse efficiencies (percent of irrigation used to increase ETa) varied from a low of 34% in the peanut experiment to a high of 89% in the oat experiment. In corn, soybeans, and sweet potatoes, the respective water-use efficiencies were 75, 53, and 51% for the better sub treatments The simulation model (NITROSIM) provided reasonable estimates of ETa as well as soil-water storage in the root zone. The model will be a useful research tool for the evaluation of water management practices under Florida soil and climate conditions. xiii

PAGE 14

INTRODUCTION The vital role of water for food production is well known in both rainfed and irrigated agriculture. In rainfed agriculture, poor seasonal distribution of precipitation and large variations from year to year may cause complete crop failure. In many humid regions, drought periods are especially detrimental on sandy soils and with crops that have shallow root systems. Hence, there has been a major growth in irrigated agriculture in many rainfed regions of the world. Irrigated agriculture, at present, is entering an age of management in which water deficits in crop production may not be totally avoided, but instead favorably controlled (Garrity et al. 1982a). There are unique differences in water management needs between arid and humid regions. In arid regions, irrigation is required to supply nearly all the crop water needs plus an additional non-evapotranspirative fraction of water to maintain a favorable salt balance in the root zone. By contrast, in humid regions, irrigation supplies less than 50% of the crop seasonal water needs and little if any irrigation is needed for maintaining salt balance. The goal of water management (irrigation) in humid regions is to maximize evapotranspirative use of the irrigation fraction of the total water input while minimizing loss of water, fertilizers, and pesticides from deep percolation beyond the root zone. Research efforts to attain this goal are only in the beginning stages. The development of irrigation management techniques which are necessary for efficient and environmentally sound uses of water, fertilizer, and pesticides requires additional research. 1

PAGE 15

2 More information is needed on irrigation scheduling (timing, frequency, and quantity per irrigation) as well as on the design and operation of irrigation systems compatible with a wide range of crop, soil, and climate requirements. The whole physical-biological system must be studied to provide basic information in the following areas: meterological and crop factors of evapotranspiration, system water balance, plant response to water stress at different growth stages, plant and soil factors influencing root system development, and root density distribution; uptake of water and nutrients, plant water stress in relation to water availability and temporal and spatial variability of soil matric potential; water and solute retention and transport in soil and plants, and drought avoidance mechanisms in plants. Irrigation strategies must be developed and tested. Supporting information should include weather forecasting, crop water production functions, risk and economic analyses, interactions of plant stresses (water, nutrients, temperature, and biological), and factors of water-use efficiency. In addition to crop production aspects of water management, there is concern about water conservation. Even in humid regions like Florida with abundant but finite water resources, state water management districts and other regulating authorities are concerned with protecting surface and groundwater quantity and quality. The present study was undertaken with the following objectives: (i) To develop evapotranspiration and irrigation production functions for oats, soybeans, peanuts, corn, and sweet potatoes under Florida climate and soil conditions. (ii) To develop economically beneficial irrigation scheduling

PAGE 16

strategies based on wetting depth, timing of irrigation, weather forecasting and crop growth stage sensitivity to water stress. The basic strategy is to replenish a part rather than the full soil water deficit in the root zone in order to maximize evapotranspirative use of water and minimize transport of water, nutrients, and pesticides beyond the root zone. (iii) To measure the water balance consequences of different water management strategies, and to use an existing simulation model (NITROSIM) to predict daily and seasonal evapotranspiration drainage, and soil-water storage within the root zone.

PAGE 17

LITERATURE REVIEW Crop Growth and Water Use There is experimental evidence that, for many crops, plant growth in terms of dry matter yield is proportional to the amount of transpiration (DeWit, 1958; Arkley, 1963; Hillel and Guron, 1973; Tanner and Sinclair, 1983). Monteith (1965) indicated that the close correlation between transpiration and dry matter yield could be explained by the fact that net radiation, which determines to a large extent the transpiration rate, and solar radiation, which determines photosynthesis, are linearly related. The stomata are the valves which allow CO^ to enter the leaf to be available for photosynthesis. However, at the same time, water loss (transpiration) occurs through the open stomates. Water stress results in a closing of stomates and a reduction in both transpiration and CO^ exchange. Since the same stomatal barrier is encountered by both and water vapor during photosynthesis and transpiration, a linear relationship between photosynthesis and transpiration would be expected (Sinclair et al. 1984). Bierhuizen and Slatyer (1965) explain the photosynthesis-transpiration relationship in terms of the vapor pressure deficit of the air. The processes are described in the following diffusion equations: T = P r +r a a (1) s P = r' +r' +r (2) a s m 4

PAGE 18

5 2 -1 where T is the transpiration rate (kg m s ) ; Ae is the vapor pressure deficit beween leaf and air; r and r are the water vapor diffusion as resistances resulting from the leaf laminar boundary layer and stomata; e is the ratio of the mole weight of water vapor to air; p and P are 3. 3. the air density and atmospheric pressure; P is the photosynthesis rate -2 -1 (kg m s ) ; ACO2 is the difference in carbon dioxide concentration of the atmosphere and at the C0_ fixation site; r' and r' are the 2 as boundary layer and stomatal resistances to CO^ diffusion into the leaf; and rm describes the CO^ diffusion resistance into the cells to the chloroplasts These research workers then expressed the photosynthesis transpiration relationship as: T, EP r' +r' +r' Ta r a s ml^ p JAe. (3) a as 2 Sinclair et al. (1984) stated that equation 1 suggests that plant transpiration rate is proportional to the vapor pressure difference between the inside of the leaf (saturated vapor pressure at the leaf temperature) and the bulk air. Arkley (1963) pointed out that the soil evaporation component of evapotranspirative water use contributes little to plant growth. Transpiration, on the other hand, is directly involved in the growth of nearly all higher plants. Thus, Arkley found that many experiments in the literature showed a highly correlated linear relationship between dry matter yield and transpiration. Plants were grown in containers where soil evaporation was prevented, and plant transpiration was corrected for mean relative humidity during the period of most active growth.

PAGE 19

6 Under field conditions, crop water use is usually expressed as evapotranspiration (ET) due to the fact that both components of ET, evaporation and transpiration, are simultaneous processes, and soil evaporation is not practically distinguishable or measured. Thus, many studies have shown that dry matter of field crops is linearly related to ET (Arkley, 1963; Hanks et al. 1969; Hillel and Guron, 1973; Stewart et al. 1977; Verasan and Phillips, 1978; Fischer, 1979). Numerous field experiments have been conducted to study the relationship between marketable crop yields and cumulative seasonal ET. In general, the relationships have been linear (Jensen and Musick, 1960; Musick et al., 1963; Stewart et al. 1977; Skogerboe et al. 1979; Hammond et al., 1981b). However, some researchers have reported a curvilinear relationship and fitted quadratic expressions to the results (Musick et al., 1976; Palacios, 1977). From the point of view of water management it is important to know under what circumstances the linear and curvilinear relationships apply. Timing of water deficits has been found to be a major cause of yield-ET variations from the linear relationship. Water deficits occurring at one growth stage might have a different effect on crop yield as compared with deficits at another stage. Thus, ET deficits during more sensitive or "critical" crop growth stages cause a relatively larger decrease in yield. Stewart et al. (1976) pointed out that when the timing of ET deficit is optimal (i.e., the minimum yield loss to be expected from any given seasonal ET deficit) the relationship bewteen yield and seasonal ET is represented by a straight line function. Sinclair et al. (1984) stated that the linearity of the response could be explained by

PAGE 20

the fact that in most of the experimental data the crop harvest Index (defined as the ratio of grain dry matter to the total dry matter yield) showed little variation over a moderate range of evapotransplration deficits. Stegman et al. (1980) reported that several researchers have found a relative Increase In cell solute concentration during the gradual development of stress In field grown plants. With this solute Increase a more negative leaf water potential develops before stomata close, and plants can tolerate more water stress before photosynthesis Is drastically reduced. Curvilinear relationships between grain yield and seasonal Irrigation are mainly due to the combination of two causes (Barrett and Skogerboe, 1980). Firstly, when water Is applied In excess of the amount required for maximum evapotransplratlon (ETm) and maximum yield (Ym) crop yields remain constant or decrease with Increased seasonal irrigation amounts. Secondly, in most cases, scientists have not reported yields with ET as the independent variable, but rather water supplied by irrigation. The yield-irrigation relationship is very sensitive to non-ET losses. When Hlllel and Guron (1973) measured the drainage component of the soil water balance, the relationship between corn grain yield and seasonal ET was strongly linear. Stewart and Hagan (1973) have well illustrated the difference In the crop yield response curves obtained with seasonal ET versus seasonal Irrigation. Stegman et al, (1980) stated that the non-ET losses were due to inefficiencies in the irrigation method as well as the inexactness Involved in scheduling irrigation according to plans.

PAGE 21

8 Other researchers have found that when irrigation was applied in amounts nearly equal to the maximum crop evapotranspiration needs, there was a linear relationship between yield and irrigation. Such linear relationships have been presented for corn (Hammond et al., 1981b), sorghum (Garrity et al., 1982a, 1982b), and soybeans (Hammond et al, 1981b; Ambak, 1982). Thus, if all of the water applied was used for the crop ET, the irrigation and ET functions would be the same. However, the irrigation function always has a smaller slope than does the ET function. The ratio of the two slopes is a measure of irrigation-use efficiency for the specific study (Stewart and Hagan, 1973; Stegman et al. 1980; Hammond et al. 1981a). Hammond et al. (1981a) pointed out that the irrigation function must be determined through research in order to develop economic water management systems and practices for particular soil-climate-crop-systems. As a concluding remark from the above review of plant growth and water use, it is obvious that crop production is linked to crop transpiration. To increase crop biomass and grain production more transpirative water needs to be used in agricultural production systems. When other growth factors are non-limiting and water amounts are adequately supplied during the crop season, the nature of the functional relationship between crop production and water use in terms of transpiration, evapotranspiration or irrigation is basically linear (Tanner and Sinclair, 1983; Sinclair et al. 1984).

PAGE 22

9 Water Management Strategies Stegman et al. (1980) stated that water management practices are generally adapted to meet the following objectives: (a) maximizing yield per unit of land, (b) maximizing yield per unit of water applied, (c) maximizing net profit, and (d) minimizing energy requirements. Thus, many researchers have developed conceptual irrigation scheduling practices which have been based on particular production systems under arid and humid climatic conditions. Hiler et al. (1974) developed the stress day index concept (SDI) for optimum timing of irrigation as follows: SDI = (SD^ X CS^) (4) where SD^ is the stress day factor which expresses the degree of water deficit in specific plant growth stage i; CS^ is the corp susceptibility factor which measures the susceptibility of plant growth stage i to a given water deficit. The crop susceptibility factor was defined as: X X. CS = i (5) X where X is the marketable yield produced by a control treatment which is kept well-watered throughout the season; is the yield in a treatment that was subjected to deficit only during the i^^ growth stage. These researchers used leaf water potential (bars) as a measure of SD^ for sorghum. They found critical SDI (SDI^) values ranging from 4 to 16 bars depending upon growth stage. The method is dependent upon experimentatal determinations of both SD. and CS for each particular crop-soil-climate system.

PAGE 23

10 Jensen et al. (1970) proposed the use of a water simulation model with allowable depletion adjusted for the stage of corp growth plus regular updating with some measure of the soil water content in the root zone. Earlier, Jensen (1968) developed a multiplicative type expression to relate the effects of limited soil water on grain yields : Ym i=l ^ETpJ. where Ya/Ym is the relative marketable yield; (ETa/ETp) is the relative total evapotranspiration during growth stage i; is the relative sensitivity of the crop to water stress in growth stage i; and n is the number of growth stages considered. The implication of this model is that the interactive effect of water stress at various growth stages on crop yield may result in some non-linearity between yield and seasonal evapotranspiration. However, limited water resources can be properly allocated by only irrigating during growth stages where yield is sensitive to water stress. Recently, Smaj stria and Clark (1981) determined values for Williams soybeans grown as early season soybeans in Florida. A plow-layer soil water management and programmed fertilization system has been developed by Rhoads (1981) for Florida Ultisols. The system involves the application of enough water to recharge the soil to plow depth before yield limiting water stress develops, and the periodic addition of plant nutrients at rates to meet the needs of plants growing at maximum rates. Rhoads obtained highest corn yields by irrigating when the soil-water tension reached -20 kPa in the plow

PAGE 24

11 layer. According to Rhoads, multiple applications of fertilizer coupled with the plow-layer water management strategy is the most practical approach to protecting fertilizer from leaching. Nevertheless, Rhoad's system might be made more efficient with due consideration of the drainage consequences of the method. Phene and Beale (1976) proposed a water and nutrient managment method in which the optimal range of soil matric potential was based on soil oxygen diffusion rate, soil strength, water desorption characteristics and unsaturated hydraulic conductivity. A high-frequency irrigation strategy was used in their studies on sweet corn. These researchers pointed out that nitrogen leaching can be minimized when soluble nitrogen and potassium are applied frequently through the irrigation system. Again, the question of drainage loss as a consequence of a nearly constant condition of maximum available water capacity is not addressed. Rawlins and Raats (1975) discussed irrigation scheduling strategies that deal with gradual depletion of available soil water during the growing season. They indicated that high-frequency irrigation goes a long way toward meeting the conflicting requirements of maintaining a high plant water potential and a sufficient capacity to store erratic rainfall. Hammond et al. (1981a) pointed out that timing, application intensity, method of application and amounts of water applied affect the fraction of added water distributed to the depletion components (ET and drainage from the root zone) and to soil water storage. Leaching losses of fertilizer and pesticides occur when there is drainage, and

PAGE 25

12 in some cases, the aeration condition of the soil root zone may be affected unfavorably during high soil water conditions. Fertilizer and water management should be coupled, especially in sandy soils where irrigation and rainfall are potentially responsible for the leaching of highly mobile ions such as N0~ (Burns, 1980). This problem has received limited research attention. Recently, Tanner and Sinclair (1983) concluded that fertility management practices in humid regions have developed with little regard to the interaction between nitrogen-use efficiency and irrigation efficiency. Thus, considerable research must be done to develop sound fertilizer/ irrigation practices which minimize drainage and leaching while maintaining an adequate water and nutrient supply for the crop. The stress degree day (SDD) concept was developed by Jackson et al. (1977) for timing irrigation. The SDD is based on the difference between the temperature of a plant canopy (T^) and the temperature of the surrounding air (T ) It has been established that as water 3. becomes limiting, T^ relative to T^ increases due to lack of water for transpiration. Thus, SDD was defined as: SDD = ? fT T > (7) n=i^ c a-^ ^ ^ where (T^ T^) is summed over N days beginning at day i. In general, if a plant is well irrigated, T T will be near zero or negative. If it is water stressed, T^ T^ will be positive. The sum of positive values will be an index of when to irrigate, i.e., water will be applied when the SDD reaches a critical value. Jackson et al. (1977) used a critival value of 10 in their studies.

PAGE 26

13 Geiser et al. (1982) developed an irrigation scheduling model by using crop-canopy-air temperature difference (AT) as the dependent variable and net radiation (Rn) relative humidity and available soil water (ASW) as independent variables. The model developed from nultiple regression analysis was: AT = -1.065+4. 71xlO^Rn+0. 02 7RH-0. 05 35ASW+4.01xlO~"'-ASW^ (8) In order to use the model for irrigation scheduling, the variable ASW was set as a constant (50% depletion) and AT was calculated for various conditions of net radiation and relative humidities. Thus, the AT value obtained from a particular Rn and RH condition represented a critical AT value (ATc) at which irrigation was scheduled. As an example of the procedure, these researchers indicated that with a net 2 radiation of 600 W/M and relative humidity of 50%, a critical value of ATc of 0.575 C was determined. As a summary of the above section, the SDI irrigation timing scheme developed by Hiler et al. (1974) follows the objective of maximizing yields per unit of water applied, especially in regions where water is a limiting resource. Both components SD and CS i i need to be determined from research for each particular crop. Jensen's model follows the objective of increasing yield per unit of water applied. However, this model has limited usefulness because of the general unavailability of the sensitivity coefficients. The plow-layer soil management and program fertilization concept developed by Rhoads (1981) for Florida Ultisols, maximizes crop yield per unit area under the assumption of an unlimited water resource.

PAGE 27

14 Even though fertilization is applied frequently to avoid leaching losses, the possibility always exists that water and fertilizer will be displaced when unexpected heavy rainfall events occur after the plow layer has been totally replenished by irrigation. The water-nutrient management method developed by Phene and Beale (1976) for humid regions maximizes crop yield and net profit per unit of water applied, and minimizes energy requirements. Because the system is based on high-frequency irrigation with a totally replenished soil profile, there is a high risk in humid regions of water loss by drainage and ground-water contamination with nitrate. The stress day degree method developed by Jackson et al. (1977) is based on a canopy temperature indicator. Although the feasibility of using thermal measurements to evaluate water stress in plants has been demonstrated, few producers have instrumentation available for timing irrigation. This method may have utility in humid regions if the effect of vapor pressure deficits and solar radiation are considered (Geiser et al. 1982). Water Use Efficiency Water use efficiency (WUE) as defined by Viets (1962) as the mass ratio of crop yield to water use. Many variations of this basic definition have appeared in literature. In quantitative terms WUE has been variously defined as the ratio of biomass accumulation, expressed as carbon dioxide assimilation, total crop biomass, or grain yield, to water use expressed as transpiration, evapotranspiration or total water input to the system. The time-scale may be instantaneous,

PAGE 28

15 daily or seasonal (Sinclair et al. 1984). Water use efficiency can be increased by (a) increasing yield without increasing water use, or (b) maintaining equal yield and decreasing water use. Yields can be increased by use of fertilizer, plant breeding, better methods of pest and disease control, and improving supplies of sunlight and carbon dioxide to the leaves. Some of these practices will invariably cause some change in the water use pattern and will directly affect WUE. Hillel and Guron (1973) state that it appears more promising to attempt to increase WUE by increasing crop yields than by decreasing evapotranspiration, since plants in the field are subject to an externally imposed evaporative demand. Viets (1966) and Black (1966) have shown that WUE is increased by improved soil fertility and cultural practices. These practices increase dry matter production without increasing ETa. However, Tanner and Sinclair (1983) and Sinclair et al. (1984) argued that WUE remains relatively constant for a particular crop regardless of improvements in soil management and cultural practices. Recently, Sinclair et al. (1984) derived an expression for seasonal evapotranspirational water-use efficiency for grain yield, WUE (ET) as follows : WUE (ET) = Hl/Kd ET/^^^ /KdE/^^] //ET. (9) a a where : H = harvest index (ratio of grain to total dry matter yield) Kd = constant incorporating plant characteristics and the energetics of C0„ conversion to biomass.

PAGE 29

16 — T — = mean daily vapor pressure deficity, weighted only for e*-e a the periods of transpiration, ET = seasonal evapotranspiration and E = seasonal evaporation from the soil. According to these researchers, the integrals of equation 9 can be eliminated with the assumption of stable conditions (Kd, E, and (e*-e) 3. are reasonably constant] during the season. Thus, equation 9 was expressed as: WUE (ET) = Cl E/ET)HKd/,— r— J. (10) (e*-e)-^ a Equation 10 shows the effect of soil evaporation on WTJE (ET) especially during the early vegetative growth when plants have not developed full canopy cover. Maximum transpirational water-use efficiency WUT(T) is reached when E is reduced to zero. As mentioned by Sinclair et al. (1984) the WUE (ET) can be improved by management practices that tend to minimize the ratio E/ET. An example practice is narrow-row spacings which result in a more rapid development of full crop canopy. Additional mecanisms that influence water-use efficiency at different scales of observation have been delineated recently by Tanner and Sinclair (1983) and Sinclair et al. (1984). These researchers gave several viable options for improving WUE such as biochemical alterations, stomatal manipulation, alteration of cropping seasons, improved harvest index and increased plant transpiration (avoiding water deficits). From the above review, one can conclude that maximum yields and high water-use efficiencies occur at the point where seasonal water use equals the evaporative demand of the atmosphere. Scheduling irrigation

PAGE 30

17 strategy plays an important role in achieving this goal, especially in humid regions where water resources are available and irrigation substantially increases agricultural production. Tanner and Sinclair pointed out that "if we are to increase national food production, the greatest increase per investment is likely to derive where productivity already is high but limited by management rather than resource" (1983, p. 24 ) And further, the authors indicated that we might best make expenditures to learn how to manage water well in the humid regions where there is water (p. 24).

PAGE 31

MATERIALS AND METHODS General From the period November 1981 to November 1982, five field water management experiments were conducted at the Irrigation Research and Education Park (IREP) on the University of Florida campus, Gainesville. Each experiment involved one of the following crops: corn, soybean, peanuts, oats, and sweet potatoes. Corn and peanut experiments were established on a Kendrick fine sand (loamy siliceous, hyperthermic family of Arenic Paleudults) which consists of fine sand material over a fine sandy loam B2t horizon beginning at a depth of 100 to 150 cm. The oat, soybean, and sweet potato experiments were established on a Lake fine sand (hyperthermic coated family of Typic Quartzipsamments) Water management treatments (rainfed and various seasonal irrigation levels) were designed to obtain a series of seasonal ET amounts in each experiment. The basic irrigation strategy was one of supplying water frequently and in amounts required to refill only a fraction of the water depleted in the root zone. Different levels of seasonal irrigation were obtained by witholding irrigation during certain droughts, varying irrigation timing and amounts, and by using portable rain shelters at selected growth stages. In the corn, soybean, and peanut experiments, water was applied at the rate of 2.54 cm from a solid-set impact sprinkler system. Quarter circle sprinklers located at the corners of 14m X 14 m plots gave a full two-sprinkler overlap along the plot 18

PAGE 32

19 edges. All four replications of a particular treatment were irrigated at the same time. In the oat lysimeter experiment measured amounts of water were applied by hand from a watering can. A portable, low pressure (20 psi) aluminum frame "Microjet" system delivered water at a rate of 1.47 cm hr to the sweet potato experiment. Plot size was 4.6 m x 6.1 m. Soil water status in all experiments was measured periodically during the growing season using gravimetric, neutron scattering and tensiometric methods. Measurements were made at 15 cm depth increments in two of the four replicationsNeutron readins (quarter-minute counting time) were obtained at least twice a week with a Troxler™ portable scaler and neutron depth probe (100 mCi Am-Be) model 1651. Gravimetric soil samples were taken infrequently only for special needs such as a more precise measurement of the water content in the 0-30 cm soil layer. Hydraulic head (H) and matric suction (h) were measured with sets of tensiometers installed adjacent (30 to 60 cm) to neutron access tubes. Irrigation timing was based on selected values of soil water suction at 15 cm depths Periodic and seasonal water depletion amounts were calculated from measured soil water contents, rainfall, and irrigation assuming no surface runoff and no upward transfer of water into the root zone. Stated in equation form: SWD = R. +1. + AS. = ETa. + D. (11)

PAGE 33

20 where : SWD^ = soil water depletion for period i (cm) = rainfall inputs for period i (cm) = irrigation inputs for period i (cm) AS_j^ = change in soil water storage in the root zone for period i (cm), ETa^ = actual evapotranspiration for period i (cm) and = drainage loss from the root zone for period i (cm) The actual evapotranspiration (ETa_j^) and drainage components of SWD. as well as soil water contents were calculated with the simulation 1 model (NITROSIM) (Rao et al. 1976, 1981). This model incorporated estimated daily potential evapotranspiration (ETp) measured soil water characteristics [field capacity (FC) permanent wilting percentage (PWP) and water redistribution time], estimated root depth with time, and a water extraction rate equal to the ETp rate until 40% of the available water (AW) in the root zone had been depleted. At that point, the extraction rate was decreased linearly with decreasing AW to zero at PWP. The Penman equation (Penman, 1948) was used to estimate daily ETp rates considering an albedo of 0.23 (vegetated surface). Official NCAA daily weather data supplied by the Department of Agronomy are not included here. Certain detailed model parameters are given in the Appendix. Root depth versus time was crop and site specific. The values used in this study were based on past experiences of researchers at the IREP and current observations of rooting depth and soil water extraction patterns. Crop coefficients were crop and treatment specific. The primary criterion

PAGE 34

21 for selection of these coefficients was current observations of the crop canopy development. This was supplemented with current leaf area index (corn) and published crop coefficients (Palacios, 1977). Soil water retention parameters were site specific and measured in the present study. Permanent wilting point was measured as the laboratory determined 15-bar water content, while the field capacity was measured (neutron and gravimetric) in the field. It was found that the model required calibration of the water redistribution parameter in order to give reasonable agreement between measured and simulated soil water contents. The water redistribution equation in the NTTROSIM model was: t FC S ^""P ^^^^ where 0 values are the volumetric water contents at time t days, field capacity, and infiltration content (maximum) for subscripts t, FC, and S, respectively. The value of a is -£n 0.01/t. In the current study, the above equation was found to redistribute the infiltrated water too rapidly. The equation was modified to t R S R ^''^^^ ^^^^ where a and t are 0,167 and 30 days, respectively. The term 9 is a R residual water content related to 0^^ and 0^^ by R = ^^FC ~ 015^/^] + (1^) where 0^^ is the volumetric water content at 15 bars water suction. These 0 values are averages for a specific soil profile and the coefficient f is defined in the following function:

PAGE 35

22 f (e^ 0^3) = (0^^ 0^3) (15) where the 0 values are average volumetric water contents as in equation 14. All modeling and statistical analyses were performed at the North East Regional Data Center (NERDC) of the State University System of Florida, Gainesville, Florida. Statistical Analysis Systems (SAS 79.5) programs were used for analysis of variance, and the Duncan Multiple Range Test was used to compare treatment means. Linear regression analyses were made with a programmable calculator. Crop response to water management in all experiments was measured in terms of total dry matter and/or marketable yields. Oat Lysimeter Experiment Oats ( Avena sativa L.) variety 'Florida 501' were planted in a 35 m X 25 m field containing 20 lysimeters (closed-bottom steel tanks 2 1.67 m in diameter, 2.2 m cross section, and 1.8 m deep). A porous ceramic plate in the bottom of the lysimeters provided for the vacuum pump extraction of drainage water to a suction of 7 kPa (Smajstrla, 1982) The soil was prepared by a rototill incorporation of a broadcast application of NH^NO^ at a rate of 60 kg ha N immediately before planting. Oat seeds were planted on 20 Nov. 1981 at a depth of 2.5 cm with approximately 22 kg of seed per ha. An additional NH^NO^ application (15 kg ha N) was made at 46 days post planting. Five water management treatments were used in a randomized complete block design replicated four times. The five treatments were:

PAGE 36

23 1. Rainfed 2. Irrigation, light-frequent, when soil-water water suction was 15 kPa at a depth of 15 cm 3. Irrigated, light-infrequent, when soil suction was 30 kPa at a depth of 15 cm A. Irrigated, medium-infrequent, when soil water suction was 30 kPa at a depth of 30 cm 5. Irrigated, light to mediuminf requent when soil water suction was 30 kPa at a depth of 30 cm. The above-ground vegetative biomass was harvested from the lysimeters on 1 Mar. 1982 as heading was beginning. The plants were dried in a forced-draft oven at 70C for several days before weighing. Corn Experiment This experiment was designed and managed by a multidiscipline team of researchers consisting of Drs. J. M. Bennett (Department of Agronomy), Dr. J. W. Jones (Department of Agricultural Engineering), and Drs. L. C. Hammond and P. S. C. Rao (Department of Soil Science). Yield data were made available by courtesy of them for crop-water production function analysis only. Detailed yield data will not be presented here. 'McCurdy 84aa' corn ( Zea mays L.) was planted on 26 Feb. 1982 in field unit 3B, IREP. Prior to planting, the land was plowed to a depth of 20-25 cm with a moldboard plow. A disk harrow incorporated 1120 kg ha of a broadcast application of a mixed commercial fertilizer: 0-10-25 (N-P^O^-K^O) plus 5.8% Mg, 0.75% Mn, 0.75% Zn, o.25% B, and 6.1% S from MgSO

PAGE 37

24 Corn seeds were planted in 60-cin rows with in-row spacing of 22 cm, resulting in a plant population of approximately 71,660 plants per ha. The experiment was a four replicate split-plot design with three irrigation treatments as main plots and two nitrogen fertilizer rates as subplots. The water management treatments were as follow: 1. Rainfed 2. Irrigated (optimal) when soil water suction reached 20 kPa at 15 cm soil depth 3. Irrigated (stress) same as treatment 2 except that a 2 to 3 week period of no irrigation for drought stress was planned to begin at 50 days. Nitrogen fertilizer treatments were: NqI Nitrogen stress during vegetative growth. Sidedress application of 27, 36, and 54 kg ha of N at 31, 35, and 67 days, respectively N^: Optimal nitrogen. Sidedress applications of 64, 52, 75, 37, 55, and 126 kg ha""*" N at 5, 20, 35, 51, 60, and 67 days, respectively. 3 Root-length densities (cm of root per cm soil) were measured in all irrigation and fertilizer treatments of replication four. Measurements were made twice during the season (93 and 103 days) Soil samples were collected in four sites of each treatment (about 5-10 cm from the stalk) Samples were collected to a depth of 120 cm in 15 cm increments with a 5-cm aluminum tube. The soil samples were washed on a sieve and roots were collected and spread on a piece of nylon mesh superimposed on a plastic sheet containing 1 x 1 cm grid lines. Root and grid line

PAGE 38

25 intersections were counted for calculation of root lengths (Tennant, 1975). Corn grain was harvested on 7 July 1982, a few days after black layer development. Peanut Experiment This experiment was designed and managed by a multidiscipline team of researchers consisting of Drs. K. J. Boote and J. M. Bennett, Department of Agronomy, and Dr. L. C. Hammond, Department of Soil Science. Yield data were made available by courtesy of them for cropwater production function analysis only. Detailed yield data will not be presented here. Peanut ( Arachis hypogea L.) seeds were hand-planted at 10 cm spacings in 60 cm rows on 5 May 1982 in field unit 3A of the IREP. Prior to planting, the land was plowed and a broadcast application (500 kg ha ^) of mixed fertilizer was incorporated with a disk harrow. The fertilizer contained 6, 25, 6.5, 5.8, 0.75, 0.75, and 0.25% of ^2*^5' ^2' ^' ^' respectively. Gypsum (CaSO^* H^O) was broadcast over the plants at a rate of 1120 kg ha on 16 June. The experiment consisted of three water management treatments (main plots) and 10 peanut genotypes (subplots) arranged in a splitplot design replicated four times. Main plots (14 m x 14 m) consisted of a total of 40 rows, with each peanut genotype occupying four rows. Reported yields will be averaged over the 10 peanut genotypes used in the experiment: UF 714021, Florigiant, UF 78114, UF 781181-1217, PI 383426, UF 79131-9, Early Bunch, UF 77318, Florunner, and UF 75102.

PAGE 39

26 Water management treatments were: 1. Ralnfed 2. Irrigated (optimal) when soil water suction reached 20 kPa at 15 cm soil depth 3. Irrigated (stress) same schedule as treatment 2, except that water stress was induced on subplots of genotypes UF 7811811217, PI 383426, Early Bunch, and Florunner during the period of 3 July to 2 August. Reported yields for this treatment will be averaged over the four genotypes. Water stress in treatment 3 was imposed by covering the subplots with portable rain shelters during expected rainfall events. On 20 Sept. 1982, a total of 6 m of row from the two center rows of each genotype were harvested by hand. Pod yields were measured and expressed in kg ha Soybean Experiment This experiment was designed and managed by a multidiscipline team of researchers consisting of Drs. K. J. Boote and J. M. Bennett, Department of Agronomy, and Dr. L. C. Hammond, Department of Soil Science. Yield data were made available by courtesy of them for cropwater production analysis only. Detailed yield data will not be presented here. 'Cobb' soybeans ( Glycine max L.) were planted on 30 June 1982 with 40 seed per m in 76 cm rows. Field unit 2 of IREP was used. The experiment consisted of a four replicate, split-plot design where six water management treatments were used as main plots with

PAGE 40

27 two induced insect defoliation levels (zero and 40-50%) as subplots. Water management treatments were: 1. Rainfed 2. Irrigated only in reproductive stages R4 and R6 when soil water suction at 15 cm soil depth reached 20 kPa 3. Irrigated as in treatment 2, but only between reproductive stages R4 and R7 4. Irrigated as in treatment 2, except that natural drought stress was permitted to develop between reproductive stages R4 and R6 5. Irrigated as in treatment 2, except that natural drought stress was permitted to develop between reproductive stages R6 and R7 6. Irrigated as in treatment 2, except that irrigation was scheduled as needed throughout the season. Soybean harvest on 4 November consisted of three rows of 3 m length each where the stand was uniform in the center of each subplot. Beans were mechanically threshed, cleaned, and weighed. Yields were expressed in kg ha Sweet Potato Experiment Sweet potato ( Convolvulus batatas L. cultivars 'Georgia Jet' and "Yellow Jewel') vine cuttings were planted on 9 July 1982 in rows 35 cm apart with 30 cm between plants. Prior to planting the experimental area was broadcast fertilized with 1120 kg ha of the fertilizer formulation used in the corn

PAGE 41

28 experiment. Two broadcast applications of 80 and 56 kg ha N in the form of NH,NO„ were made at 17 and 41 days after planting, respectively. 4 i The experiment consisted of a four replicate, split-plot design with six water management treatments as main plots and two varieties as subplots. Water management treatments were: 1. Rainfed 2. Irrigated (optimal) when soil water suction reached 20 kPa at 15 cm soil depth 3. Irrigated, when soil water suction reached 20 kPa at 30 cm soil depth 4. Irrigated, when plants exhibited visible wilt 5. Irrigated (stressed) same as treatment 2, except that water stress was induced during the period 40 to 70 days after planting 6. Irrigated (stressed) same as treatment 2, except that water stress was induced during period 70 to 100 days after planting. Water stress periods in treatments 5 and 6 were imposed by covering the plots with portable rain shelters (40-70 and 70-110 days, respectively) only when rain was expected. 'Georgia Jet' and 'Yellow Jewel' sweet potato varieties were harvested on 4 and 15 Nov 1982, respectively. Potato tuber fresh weights were obtained from the two central rows of each subplot by 2 taking an area of 4.0m The tubers were graded according to commercial specifications, and subsamples were taken directly to a forceddraft dryer for several days and then weighed to obtain potato dry

PAGE 42

29 2 matter yields. Above-ground vegetation was measured from 1.70 m 2 of the 4.0 m used for harvestable yield. The whole sample was dried at 70C. In both varieties, root length densities were measured at harvest time in treatment 1 (rainfed) and treatment 2 (well-irrigated) Root samples were obtained in a single site (about 10 cm from the main stem) per subplot of replications 1, 2, and 3. The soil-root samples were taken in 15 cm increments to a depth of 210 cm with a 5 cm aluminum tube. Roots were separated from the soil as described for the corn experiment. Root length densities were obtained by using a gridintercept method (Tennant, 1975).

PAGE 43

RESULTS AND DISCUSSION Oat Lysiineter Experiment Rainfall and Irrigation Seasonal rainfall distribution and irrigation schedules for the growing season of 'Florida 501' oat are shown in Fig. 1 and Table 1, respectively. Rainfall was regularly distributed over the season, however, five dry periods with durations of 10 to 24 days were observed. Two heavy rainfalls on days 55 and 88 (8.0 and 6.7 cm) caused water losses by drainage in all treatments. Irrigation was used only between days 75 and 97 (3 through 25 February). Evapo transpiration Periodic and seasonal evapotranspiration rates were obtained from water balance computations. Because periodic actual evapotranspirations for all treatments were nearly the same during the period 0-75 days, average ETa values were used. Figure 2 shows a comparison between ETa rates for treatments 1 and 3 (rainfed and well-irrigated) and the potential evapotranspiration rate (ETp) as estimated by Penmans's (1948) method. Irrigation in treatment 3 increased the ETa rates over the ETp rates furing the irrigated period (75 to 97 days). In contrats, the ETa rates of the rainfed treatment almost reached the ETp rate during the same period of time. Note that the ETa for both treatments were lower than the ETp rates from the period of 0 to 60 days. During 30

PAGE 44

31 PLANTING, N0V.20, 1981 TREATMENT 5 HARVEST.MARCH 1, 82 TREATMENT 4 TREATMENT 3 10 TREATMENT 2 TREATMENT 1 1 20 40 60 80 DAYS FROM PLANTING 100 110 Fig. 1. Rainfall distribution and irrigation schedules for 'Florida 501' oat dry raatter production.

PAGE 45

32 Table 1. Irrigation schedule, 'Florida 501' oat, 1981-1982. Irrigation amount on treatment number Date Day 2 3 4 5 cm Feb. 3 75 1.3 1.5 5 77 1.5 6 78 0.5 7 79 2.5 1.3 11 80 1.0 — — — 13 85 1.5 15 87 1.0 2.5 20 92 1.5 21 93 1.0 2.0 24 96 1.0 25 97 1.5 TOTAL 5.8 6.0 4.5 5.3

PAGE 46

33 H rO 4J 73 -4 o 3 c • 00 CO > 00 1— I CT\ •H fH CO •V 01 CO 6C O to U > O CO us 'V CO 0) U -H 3 !-i 0) O CO iH 01 fci X (l_ Aop ujuj ) 13 6C •H

PAGE 47

34 the early part of the season, incomplete canopy cover could not totally intercept solar radiation, so ETa rates were lower than the potential rates. When the crop develops a leaf area index of 4 almost 90% of the solar radiation is intercepted (Ritchie, 1972) and ETa rates reach or exceed ETp expecially when there is advection of energy from non-irrigated areas. The water-use coefficients (Kc) for oat dry matter production were calculated assuming that oat plants in the early part of the season were growing with no severe stress. Experimental ETa data of well irrigated treatment 3 and estimated ETp values were related in the following expression: Kc ETa (^) (16) Figure 3 shows the variation of Kc coefficients over the growing season. Crop coefficients eventually exceeded 1 at the end of the growing period. At harvest, the oat plants had developed a full canopy, and were in transition from vegetative to reproductive stages. Advection of energy from unirrigated areas surrounding the lysimeter could have accounted for the increase in ETa over ETp. Dry Matter Yield Response Irrigation amount, seasonal evapotranspiration and oat dry matter yields for all treatments are summarized in Table 2. All irrigated treatments produced higher yields (0.01 level) than the rainfed treatment. The highest dry matter yield was obtained in irrigated treatment 3.

PAGE 48

35

PAGE 49

36 Table 2. Effect of water management on dry matter yield of 'Florida 501' oat, 1981-1982. Treatment! Irrigation ETa Dry matter yield cm kg ha 1 0 20.50 1450 d§ 2 5.80 (6)t 25.70 4300 ab 3 6.00 (4) 26.00 4520 a 4 4.50 (2) 24.60 3610 c 5 5.30 (3) 25.00 4050 b fSee p. 23 for description of treatments. ^Numbers in parenthesis are the number of irrigation applications. §Means followed by the same letter are not significantly different (0.05 level, Duncan's Multiple Range Test).

PAGE 50

37 The relationship between dry matter yield and seasonal evapotranspiration as well as irrigation was determined by simple linear regression analysis. The data were plotted (Fig. 4) as suggested by Stewart and Hagan (1973). Dry matter yield was linearly related to ETa and irrigation. Note that the two linear relationships are not superimposed since a given irrigation amount did not produce an equal increase in evapotranspiration. The dry matter-ETa relationship conforms with the findings of several researchers (Arkley, 1963; Hanks et al., 1969; Hillel and Guron, 1973; Stewart et al. 1977; Fischer, 1979). It is more common to find a curvilinear (concave downward) relationship between yield and irrigation (Yaron, 1971; Shipley and Regier, 1975, 1976). Stegman et al. (1980) indicated that the irrigation function is typically curvilinear and curves away from the ET function as irrigation amounts increase. The yield-irrigation linear relationship obtained in the present research can be explained in terms of water management factors. The irrigation rates in each irrigated treatment were not high enough to produce large drainage, i.e., drainage losses were proportional to irrigation amounts. In addition, nutrient losses or root aeration problems were not observed. The regression coefficient of both dry matter-ET and dry matterirrigation relationships are called crop-water production functions. The irrigation production function is less than or equal to the ET production function. The goal in water management is to obtain an irrigation function as near the ET function as possible.

PAGE 52

39 The ET and irrigation functions obtained in this research were 556 and 499 kg ha cm respectively. The ratio of these two values (499/556 = 0.90) provides an estimate of the fraction of the water applied that was used by the crop as ET (Hammond et al. 1981a). Thus, the smaller the slope of the irrigation function the less efficient was the water management strategy during the season. The rather large seasonal water-use efficiency of 90% could be attributed to a good combination of irrigation scheduling and rainfall. Figure 5 shows the dry matter and ET data plotted as yield reduction in response to decreasing seasonal ET. The slope of this relationship (Ky) called the yield response factor (Doorenbos and Kassam, 1979) gives the unit of relative dry matter reduction per unit of relative ET deficit. Thus, from this relationship, it is observed that a seasonal ET deficit of 20% will produce a 64% reduction in dry matter production (3.20 x 0.20 = 0.64). A relative ET deficit of 20% represents an absolute ET deficit of 5.2 cm (0.20 x 26 cm). A relative dry matter yield decrease of 64% represents an absolute yield decrease of 2893 kg ha ^ (0.64 x 4520 kg ha ^) Dividing the absolute dry matter decrease by the absolute ET deficit (2893/5.2) gives 556 kg ha"""" cm'"*", which is the value of the ET production function previously obtained (Fig. 4). Water Balance Measured periodic water balance values for the oat experiment are presented in Table 3. Total water balance for the growing season is summarized in Table 4. Water depletions were calculated from

PAGE 53

Fig. 5. Relationship betx^een oat relative dry matter yield decrease (1 Ya/Ym) and the relative seasonal ET deficit (1 ETa/ETm)

PAGE 54

Al Table 3. Periodic measured water balance during the vegetative growth period of 'Florida 501' oat, 1981-1982. Treatmentt Inputt as Depletion Drainage ETa§ cm 0-6 days (ETp = 1.48 cm)// i-J 11 f-\ 1 O "7 -0 lo 7 0 441 0 133 0. 308 6-32 days (ETp = 5.16 cm) J. J — U JfO J 0 J 1 AQ 1 "7 c: 32-44 days (ETp = 2.10 cm) 1-5 2.69 +0.77 1.92 0.17 1.75 44-46 days (ETp = 0.39 cm) 1-5 0.0 -0.36 0.36 0.0 0.36 46-57 days (ETp = 1.90 cm) 1-5 10.62 +1.95 8.67 7.31 1.36 57-64 days (ETp = 1.30 cm) 1-5 0.28 -1.61 1.89 0.0 1.89 64-75 days (ETp = 2.58 cm) 1-5 2.82 -1.83 4.65 0.0 4.65 75-95 days (ETp = 5.61 cm) 1 9.88 -1.24 11.12 5.45 5.67 2 14.13 -1.00 15.13 5.84 9.29 3 13.88 -0.74 14.62 5.31 9.31 4 13.88 -1.29 15.17 6.00 9.17 5 14.88 -0.79 15.67 6.17 9.50 95-98 days (ETp =0.98 cm) 1 0.0 -1.24 1.24 0.0 1.24 2 1.0 -0.82 1.82 0.0 1.82 3 1.50 -1.45 2.95 1.14 1.81 4 0.0 -1.15 1.15 0.0 1.15 5 0.0 -1.25 1.25 0.0 1.25

PAGE 55

42 Table 3 — continued. Treatmentf InputJ aS Depletion Drainage ETa§ cm 98-101 days (ETp =0.83 cm) 1 0.0 -0.10 0.10 0.0 0.10 2 0.0 -1.53 1.53 0.0 1.53 3 0.0 -1.82 1.82 0.0 1.82 4 0.0 -1.18 1.18 0.0 1.18 5 0.0 -1.21 1.21 0.0 1.21 tSee p. 23 for description of treatments. $Rainfall plus irrigation. §ETa obtained as the difference of measured soil water depletions and measured drainage. Upe riodic water balance averaged for all treatements from the period of 0-75 days. ^^From daily estimates, Penman method. See Appendix Table 27.

PAGE 56

H m m m m n m CN CN CN est (N CM CSI CN H W o o o m vD in CN CN CN CN CN (U 60 (0 c •H n) P 00 m CO in in iH CO m CO tX5 CO CO in 00 o
PAGE 57

44 neutron readings taken at periods of time given in Table 3. As mentioned earlier, because irrigation was used only in the last part of the season (from day 75 to day 97) an average (over treatments) of periodic water balance values was computed for the 0-75 days period. Periodic actual ET values were calculated as the difference between periodic depletion (Table 3) and periodic drainage. Large drainage losses (90% of the season total) occurred following two heavy rainfalls (days 55 and 85) Average per treatment irrigation input was 5.4 cm; average ETa increase was 4.83 cm or 4.83/5.4 = 0.89 efficiency average drainage increase was 1.16 cm, but some of this came from a decrease in stored water = average of 0.59 cm. Thus, the average increase in drainage attributed to irrigation was (1.16 0.59) 0.57 cm or 0.57/5.4 = 0.11. Hence, 89% of irrigation contributed to an increase in ETa and 11% went to an increase in drainage. The 89% irrigation-use efficiency compares favorably with the 90% efficiency calculated by the ratio of the two production functions. Corn Experiment Rainfall and Irrigation The rainfall and irrigation distributions for the growing season of the 'McCurdy 84aa' com are shown in Fig. 6 and Table 5. All treatments received an initial irrigation of about 1.27 cm on 4 March which was recorded as rainfall. The maximum daily rainfall, 9.88 cm, was recorded on day 42. A rain-free period between 55 and 75 days permitted

PAGE 58

45

PAGE 59

46 Table 5. Irrigation schedule, 'McCurdy 54aa' corn, 1982, Irrigation amount on treatment number Date Day 2 3 cm March 18 21 1.02 1.02 April 8 42 1.02 1.02 17 51 1.27 1.27 21 55 1.78 1.78 May 2 66 2.54 6 70 2.54 9 73 1.27 12 76 1.52 13 77 3.05 14 78 1.90 0.38 17 81 2.03 2.03 21 85 2.03 2.03 30 85 2.03 2.03 June 7 102 1.90 1.90 10 105 0.63 0.63 11 106 2.54 2.54 TOTAL 25.51 19.17

PAGE 60

47 the development of a planned 2-week water stress period for treatment 3 Differential irrigation treatments began on day 21 when treatments 2 and 3 were irrigated. Thereafter, irrigations were accomplished for the two irrigated treatments when the required conditions were met. The total seasonal amount of irrigation applied in treatments 2 and 3 were 25.5 and 19.2 cm, respectively. Yield Response Dry matter yield response to irrigation and estimated ETa fit linear functions very well (Figs. 7 and 8). In contrast, grain yield response gave a good fit only under nitrogen sufficient conditions (Figs. 9 and 10). For dry matter, nitrogen stress reduced irrigation and ET regression coefficients by 24 and 39%, respectively. The associated reductions for grain yields were 57 and 46%. The grain ET function under nitrogen sufficient conditions (457 kg ha""'' cm""*") is in agreement with calculated values from corn grain data of Hillel and Guron (1973) in Israel (540, 440, and 450 kg ha'"*" cm""*") However, Hammond et al, (1981b) reported a larger ETa production function (661 kg ha cm "'') Although treatment 3 was water stressed during a critical growth stage, there was no significant departure from linearity in any of the four functions in Figs. 7 and 9. However, there were only three water management levels, and the rainfed treatment was a rather severe drought treatment. Tanner and Sinclair (1983) and Sinclair et al. (1984) state that the relationships between grain yields and ET are essentially

PAGE 61

48 OS OS O O II II CN C£ ^ — s CO c H h-l W Di M CM CN vO + + CM rH Cd hJ •a cn t-i c OJ c H •H iJ iJ •H c ij . -H u U CJ CJ c 3 tc •H cn • 0) ^ 4-1 -H >-l 4J 3 4J O 3 O 5"

PAGE 62

e o o pco H O < > < o CO < I CO OO •H M eS C o m rt 0) 0) c w H W .-I CNI r3 00 C a^ O .H cn C3 QJ CO cn c o cn -H 3 w cn -H 0) > c o CJ T3 1— I cn cu cn •H OJ QJ cn 4J to e c (U 60 O S-i w 5-4 -H T3 C d !-i o -a o c 3 C3 -ac CO M >, ci t3 M !-i 3 U C o o 2^ -H 00 o 00 •H

PAGE 63

50 CO c O CM CD 00 0^ 0) ^ CO -a X c c ca o •H CB 4-1 H -H W Tl C -H O cfl O c o -u tn c CS 01 0) -H to CJ •H cn M-i 3 14-1 cn > 3 CO c CiC o •H c c •H U CO QJ !-i -a M C 3 c o o O M CO M CO w 00 c o >. -H ( gOix j_Dq 5>i) a-i3i>, Nivao 3 u o 0^ •H CO M •H u

PAGE 64

51 o I— I c o 03 C O cn c CO CO H TO C o cn Cfl M CM 3 00 S-J r-( > 1 — ( X c o C C •H O TO O tC Cfl C OJ !-i O -U U CO C C3 00 O i-i 4-1 -H -a c a -a S c 3 (JC

PAGE 65

52 linear due to the fact that crop harvest indexes remain relatively constant over levels of ET deficits. Curvilinear (concave downward) relationships between grain yield and irrigation are commonly found when treatments receive excessive irrigation (Stegman et al., 1980), Yield reductions at high levels of irrigation can result from poor soil aeration and leaching of plant nutrients Note that the irrigation production coefficients were lower than the ET production coefficients (Figs. 7, 8, 9, and 10). The ratio of these coefficients (irrigation/ET) provides an estimate of the fraction of the seasonal irrigation which was used to increase evapotranspiration. Thus, these ratios represent the average irrigation-use efficiency achieved in the specific experiment. The efficiencies were higher under nitrogen sufficient conditions (0,78 and 0.75 for dry matter and grain responses, respectively). The respective efficiencies under nitrogen stress were 0,62 and 0,60, Hammond (1981b) stated that the goal in irrigation management is to achieve a ratio equal to one. This ideal is not achieved very often, because some of the irrigation may remain stored in the root zone or rainfall may displace it from the root zone. The data from Figs, 9 and 10 were replotted in Fig. 11 to more clearly show the interaction of nitrogen and water stress. The relative corn grain yield reduction as a function of relative ET deficit is shown in Fig. 12 for N sufficient conditions. The slope (Ky) of this relationship, called the yield response factor by Doorenbos and Kassam (1979) indicates the unit of relative yield reduction per unit of relative ET deficit. Thus, under rainfed conditions (treatment

PAGE 66

T r 120007 10000o 5* 8000. o -J y 6000. > Z < 4000(T O Y = -8882 + 457 (ET) R = 0.93 NO -2812 + 248 (ET) 2000 R = 0.83 4 28 32 36 40 44 48 52 SEASONAL EVAPOTRANSPIRATION (cm) SEASONAL IRRIGATION (cm) 11. 'McCurdy 84aa' corn evapotranspiration and irrigation production function under nitrogen sufficient (N ) and nitrogen stress (N ) conditions, 1982.

PAGE 67

54 •H U a H w (U rH 0) ^ CD 3 CO M 0) cn > C o e 4-1 CO C trt O a I u H C w ^ H K a O -H •H U-1 CJ 3 3 cn -a (U c M dj ec a o >^ C C ^1 •H (1) cd -a M c OC 3 C /-V ^1 6 C H CJ Cd CO I 01 Pi CM •H

PAGE 68

55 1) the relative seasonal ET deficit of 43% produced a relative grain yield reduction of 74% (1.72 x 0.43 = 0.74). The relation of Ky to the production coefficient is given by the following expression: ET = Ky (Ym/ETm). (17) coeffxcient Thus, for the corn data in Fig. 6, we have ET ^ = 1.72(12300 kg ha"''"/46.35 cm) = 457 kg ha~"'"cm~''". coefficient v e. / & Root Length Density Distribution Roots of 'McCurdy 84aa' corn were sampled on day 103 for all treatments. The samples were taken in four sites for each treatment of replication IV. Root length density distributions are given in Fig. 13. There was a significant water management x nitrogen x depth interaction. No other factors were significant. The rainfed nitrogen stress treatment (WING) had the lowest concentration of roots in the 0-15 cm and 30-60 cm layers. There were no differences in root density distributions in the 50-105 cm layer, but there were differences in the 105-120 cm layer. No roots were found in the well irrigated treatment ( 2 ) and lower root density values were found with nitrogen stress than with sufficient nitrogen in both the rainfed and limited irrigation (3) treatments. Apparently, the stunted growth resulting from a combination of water and nitrogen stress reduced photosynthate production and delivery to the root system. In addition, root penetration of the soil profile could have been restricted due to increasing soil strength with drying. Persad (1982) found that root distribution of corn plants, in the same area in 1981, was restricted by the presence of a tillage pan at the 30-50 cm depth.

PAGE 69

ROOT LENGTH (cm cm^ of soil) 13. Root length density distribution of 'McCurdy 84aa' on day 103 after planting, 1982.

PAGE 70

57 Water Balance Periodic water balance data showing a comparison between measured and simulated profile water depletions for 'McCurdy 84aa' corn are presented in Table 6. The seasonal water balances calculated by the summation of the measured or estimated periodic water balance data are summarized in Table 7. Measured periodic water depletions in Table 6 were calculated from neutron readings. Estimated ETa, drainage, and water depletion were obtained by simulation. As shown in Table 6, large drainage losses occurred in the first 43 days of the growing season (63, 55, and 58% of the total estimated drainage in treatments 1, 2, and 3, respectively). In general, there was close agreement between measured and estimated periodic depletions for both nitrogen conditions. However, lack of agreement was particularly observed in periods 75-81 and 81-84 days for the well irrigated treatment 2 under nitrogen sufficient conditions. Measured water depletion for the first period (75-81 days) was 0.42 cm while the estimated depletion was 3.55 cm. This indicates that for measured depletion most of the water input was stored in the soil, and that the combined ETa and drainage was only 0.42 cm for the 6-day period. This, of course, is not reasonable because the simulated ET for the same period was 3.43 cm. In contrast, the measured and estimated depletions for the next period (81-84 days) were 5.96 and 1.91 cm, respectively. For the measured case it would have to be a considerable drainage event since the estimated ET value was only 1.69 cm. Thus, the change from low to high depletion is not reasonable and perturbations in the measured

PAGE 71

58 Table 6. Periodic water balance during the growth period of 'McCurdy 84aa' corn, 1982. Measured water Estimated water Treatmentt Input"^" AS^ Depletion ETa Drainage AS^ Depletion' cm 0-19 days, N and N (ETp =6.90 cm) 1-3 7.04 0.02 7.02 1,85 5.64 -0.45 7.49 19-40 days (ETp = 7 ,78 cm) 1 8.01 0.16 7.85 3.50 4.89 -0. 38 8. 39 2 9.04 0.11 8.93 4.56 4.96 -0.48 9.52 3 9.04 0.16 8.88 4.53 5.00 -0.49 9.53 19-43 days, Nq (ETp = 8 65 cm) 1 23.05 2.78 20.27 3.88 16 82 2. 35 20 70 2 25.10 2.72 22.38 5.06 17.69 2.35 22.75 3 25.79 2.81 22.98 4.98 18.44 2.37 23.42 40-43 days (ETp = 0, .87 cm) 1 15.04 2.81 12 23 0 59 11 70 9 7S 19 9Q 2 16.06 2.70 13.36 12.23 3.09 12.97 3 16.06 2.75 13.31 0.69 12.15 3.22 12.84 43-57 days, Nq (ETp = 6. ,06 cm) 1 4.38 -5.56 9.94 3.11 5.18 -3.91 8.29 2 7.43 -4.10 11.53 4.93 5.12 -2.62 10.05 3 7.43 -4.00 11.43 k.ll 5.16 -2.50 9.93 43-57 days (ETp = 6. 60 cm) 1 4.38 -5.96 10.34 3.30 5.04 -3.96 8.34 2 7.43 -3.88 11.31 5.35 5.00 -2.92 10.35 3 7.43 -4.12 11.55 5.15 5.06 -2.78 10.21 57-65 days N-j^ (ETp = 3. 06 cm) 3 2.76 -1.49 4.25 2.84 0.24 -0.32 3,08

PAGE 72

59 Table 6 — continued. Measured water Estimated water Treatmentt Input^ ^S§ Depletion ETa Drainage as§ Depletion^ 57-68 days, (ETp =4.43 cm) 2 5 .30 u uy 5 zl 4.00 0.43 0 .87 4 .43 3 2 .76 -1.58 57-68 4.34 days, 1 3.43 (ETp =4.43 0.54 cm) -1 .21 3 .97 2 5 .30 0. 74 57-71 4.56 days, Nq 4.51 (ETp =5.90 0.37 cm) 0 .42 4 .88 1 2 .87 -1. yy 4 Ob 3.21 0.05 -0, .39 3 .26 57-71 days (ETp =5.90 cm) 1 o .87 -1.92 0 J-DO 4.79 days 3.27 (ETp =1.37 0.11 cm) -0, .51 3 .38 •3 J n u -1.20 £ Q "71 DO/ 1 1.20 days, Ng 0.88 (ETp = 1.47 0.01 cm) -U sy (J 9 Z 1.30 1.34 1.41 0.54 U by 1 95 3 0, .10 -1.31 68-71 1.41 days 0.69 (ETp =1.47 0.04 cm) -0, .63 0, 73 Z 0 z 04 1.06 1.58 1.61 0.36 0. 67 1, .97 3 0, .10 -0.46 0.56 0.66 0.02 -0. .58 0. .68 71-73 days, Nq (ETp =0.99 cm) o o U -0.15 71-73 0.15 days 0.38 (ETp =0.99 0.09 cm) -0. 38 0. 38 1 0 -0.94 0.94 0.30 0 -0. 30 0. 30 3 0 -0.90 0.90 0.37 0 -0. 37 0. 37 71-74 days, Nq( [ETp = 1.53 cm) 2 1. 27 -1.87 71-74 3.14 days 1.47 (ETp = 1.53 0.29 cm) -0. 49 1. 76 2 1. 27 -1.51 2.78 1.68 1.04 -1. 45 2. 72

PAGE 73

60 Table 6 — continued. Measured water Estimated water Treatmentt Inputt as§ Depletion ETa Drainage aS§ Depletionlf cm 71-76 days, (ETp = 2.53 cm) 1 0 -1.49 1.49 0.72 0.02 -0.74 0.74 73-74 days, (ETp =0.54 cm) 3 0 -0.20 0.20 0.31 0 -0.31 0.31 73-76 days, Ng(ETp = 1.54 cm) 3 3.05 2.02 1.03 0.78 0.20 2.07 0.98 7376 days, (ETp =1.54 cm) 1 0 -0.32 0.32 0.38 0.01 -0.39 0.39 7475 days, (ETp = 0.53 cm) 2 1.52 -1.31 2.83 0.50 0.09 0.93 0.59 74-76 days, (ETp =1.00 cm) 3 3.05 2.36 0.69 0.48 0.35 2.22 0.83 7481 days, Ng(ETp =3.55 cm) 2 5.46 2.18 3.28 3.37 0.58 1.51 3.95 7581 days, N^(ETp =3.05 cm) 2 3.94 3.52 0.42 3.43 0.12 0.39 3.55 7681 days, Nq(ETp =2.55 cm) 1 0 -0.71 0.71 0.51 0 -0.50 0.50 3 2.41 0.90 1.51 2.23 0.03 0.15 2.26 76-81 days, (ETp =2.55 cm) 1 0 -0.73 0.73 0.47 0 -0.47 0.47 3 2.11 0.03 2.08 2.31 0.01 -0.21 81-82 days, (ETp =0.47 cm) 2.32 -0.89 0.89 0.43 0 -0.43 0.43

PAGE 74

61 Table 6 — continued. Measured water Estimated water Treatmentt Input^ aS§ Depletion ETa Drainage aS§ Depletion cm 81-84 days, N (ETp =1.47 cm) 2 2. ,03 -2.66 4.69 1.38 0.76 -0.11 2.14 J 2. U J -0.60 2.63 1 J U • U J u 1 1/. 1 j'f 81-84 days Nj^ CETd = 1 47 cm) 1 0 -0.01 0.01 0.25 0 -0.25 0.25 / U J -3.93 5.96 n 0 0 U ZZ U iZ 1 Q 1 81-88 days, CETd = 3 28 cm) 1 6. ,32 2.96 3.36 1.25 0.13 4.94 1.38 82-84 days (ETp =0.99 cm) 3 2. ,03 -0.56 2.59 0.89 0.25 0.89 1.14 84-88 days, Nq (ETp =1.81 cm) 2 6. 31 3.73 2.58 1.70 2.64 1.97 4.34 3 6. 32 3.88 2.44 1.66 0.22 4.17 2.15 84-88 days (ETp = 1.81 cm) 1 6. 32 3.27 3.05 1.01 0 5.19 1.13 2 6. 31 4.42 1.89 2.08 1.44 1.97 4.34 3 6. 32 4.71 1.61 1.73 0 4.42 1.90 88-91 days, Nq (ETp = 1.26 1 cm) 2 0. 53 -1.77 2.30 1.19 1.65 -2.28 2.81 3 0. 52 -2.69 3.22 1.17 1.67 -2.13 2.66 88-81 days (ETp =1.26 cm) 1 0. 53 -0.56 1.03 0.88 0 -0.49 1.02 2 0. 53 -1.99 2.52 1.45 1.55 -2.44 2.97 3 0. 53 -0.94 1.47 1.22 1.56 -2.13 2.66

PAGE 75

62 Table 6 — continued. Measured water Estimated water Treatmentt Input^ AS§ Depletion ETa Drainage aS§ Depletion^t cm 88-98 days, N (ETp =4.19 cm) 1 5.18 1.37 3.81 2.62 1.46 1.10 4.08 91-98 days, Nq (ETp = 2.93 cm) 2 6.17 0.32 5.85 2.58 3.88 -0.27 6.34 91-98 days, (ETp =2.93 cm) 1 4.65 1.63 J \j ^ -L./D X.jj 1.25 3.40 2 6.17 0.08 6.09 2.93 3.40 -0.13 6.30 3 6.17 0.98 5.19 2.57 3.83 -0.23 6.40 91-111 days, Nq (ETp =9.98 cm) 3 17.13 0.32 16.81 8.15 9.94 -0.87 18.10 98-102 days, (ETp =2.20 cm) 1 0 -0.41 0.41 1.32 0.87 -1.70 1.70 98-105 days, (ETp =3.95 cm) 2 5.08 -2.07 7.15 3.89 0.58 0.32 4.76 98-111 days, Nq (ETp =7.05 cm) 1 5.89 1.90 3.99 4.00 2.49 -0.61 6.50 2 10.97 2.06 8.91 5.90 5.76 -0.71 11.68 98-111 days, (ETp =7.05 cm) 3 10.96 -0.38 11.34 5.91 5.75 -0.71 11.67 102-111 days, (ETp =4.84 cm) 1 5.89 2.38 3.51 2.68 1.60 1.55 4.34 105-111 days, (ETp =3.10 cm) 2 5.89 0.46 5.43 2.79 4.44 -0.83 6.72 111-124 days, (ETp =5.73 cm) 1 14.60 1.23 13.37 2.77 11.39 0.45 14.15 2 14.61 1.97 12.64 4.26 10.03 0.33 14.28 3 14.61 0.44 14.17 4.21 10.11 0.31 14.30

PAGE 76

63 Table 6 — continued. Measured water Estimated water Treatmentt Input! as§ Depletion ETa Drainage as§ Depletionlf cm 111-124 days, (ETp =5.73 cm) 1 14.60 0.66 13.94 2.97 11.29 0.38 14.22 2 14.61 2.18 12.53 4.83 9.55 0.22 14.39 3 14.61 0.69 13.92 4.26 10.03 0.30 14.31 124-127 days, (ETp =1.44 cm) 2 0 -1.18 1.18 1.01 0.49 -1.47 1.47 3 0 -2.06 2.06 1.01 0.49 -1.48 1.48 124-127 days, (ETp =1.44 cm) 1 0 -1.38 1.38 0.72 0.78 -1.49 1.49 2 0 -1.05 1.05 1.15 0.32 -0.95 0.95 3 0 -0.86 0.86 1.01 0.50 -1.49 1.49 124-130 days, (ETp =3.02 cm) 1 0.41 -1.35 1.76 1.33 1.18 -2.08 2.49 127-130 days, (ETp 1.58 cm) 2 0.41 -1.91 2.32 1.11 0.03 -0.73 1.14 3 0.41 -0.50 0.91 1.11 0.03 -0.73 1.14 127-130 days, (ETp =1.58 cm) 1 0.41 -0.60 1.01 0.79 0.25 -0.63 1.04 2 0.41 -0.50 0.91 1.26 0.01 -0.87 1.28 3 0.41 -0.41 1.24 1.10 0.02 -0.71 1.12 tSee p. 23 for description of treatments. ^Rainfall plus irrigation. §Soil profile water content on last day of period minus content on first day. IfEstimated ET plus estimated drainage.

PAGE 77

D4J w COS o M •H CM o •H n 00 CTl 4J CO CO 0 C H 0 o ^ — ^ •H ro CO o o w QJ ro tn CO ^ QJ (50 00 rH CO Q CM CM — 1 — 1 & 1 1 1 1 I QJ QJ •rf rH T3 CO V4 CO a E-H •H o rH 00 vD in CT. CO 00 CTN u 14-4 3 CO 73 CO QJ
PAGE 78

65 water contents must represent instrument and/or operator errors rather than underestimation or overestimation of both ETa and drainage. It is obvious that other factors which should influence the results in a more consistent manner include spatial variation in irrigation amounts, temporary water storage as a result of variable clay layer location, variable length of period between soil water measurements, and variation in precision of the neutron-water content relationship with water content in the 0-30 cm soil layer. Additional factors are the assumptions associated with the simulation model: estimated ETa, crop coefficients, water storage capacity, and residence time of water in the root zone. Differences between measured and estimated seasonal depletion varied from nil (treatment 1, N^) to a maximum of 2.3 cm (treatment 2, N^) The measured input was subject to error. Irrigation was spatially variable and the pattern changed with wind velocity and direction. The measured change in soil storage was based on a limited neutron sample of the spatially (three dimensions) variable soil water content. The difference (AS) between measured water contents at the beginning and end of a period plus the measured water input gave the measured depletion. Simulated depletion resulted from the model allocation of measured inputs among ETa, drainage, and change in stored water. The sum of these ETa and drainage components equaled simulation depletion. Thus, with the same input base, depletion differences (measured and simulated) must equal differences in the associated AS values. It follows that the basic parameter for comparison in these measurements and simulations is the soil water content or depth in the root zone.

PAGE 79

66 Figure 14 shows the measured and simulated root zone water content distributions with time for all treatments under nitrogen sufficient conditions. Simulated water contents were lower than measured values for all treatments during the first 60 days of the growing season. From day 60 to the end of the season close agreement was obtained in irrigated treatments 2 and 3. The data in Fig. 14 suggest that the model alone could be used successfully in scheduling irrigation. Measured and simulated water content distributions for all treatments under nitrogen stress conditions are presented in Appendix Table 28. The pattern of measured and simulated values under nitrogen stress was similar to that shown in Fig. 14 for nitrogen sufficient conditions The periodic and seasonal changes in soil water content (AS) are of interest, too. The seasonal values contain a built-up initial water content profile consisting of soil depth segments added as root depth increases to maximum. It follows that measured and simulated profiles will be different since they contain the differences in measured and estimated water contents in each of these added soil layers. The dynamics of the system make each layer unique in terms of the water status at a particular instant. Water contents predicted by the model and those measured by the neutron meter may agree better under some circumstances than others. In view of all the factors involved, the present limited test of validity of the model is very encouraging. The actual evapotranspiration (ETa) component of estimated periodic depletions for treatments 1 (rainfed) and 3 (stressed) were

PAGE 80

67 Fig. 14. Measured and simulated root zone water content distribution with time for three water management treatments, under nitrogen sufficient conditions. 'McCurdy 84aa' corn, 1982.

PAGE 81

68 lower than the periodic ETp (Penman, 1948) under both nitrogen conditions throughout the growing season (Table 6). However, the estimated ETa for the well irrigated treatment 2, under nitrogen sufficient conditions equaled or exceeded the ETp rates for period 57-68 to 91-98 days. This resulted from the selection of crop coefficients (Appendix Table 23) equal to or greater than 1 during mid-season. Thus, we should find ETa values greater than ETp during those periods. Even though estimates of ETa were higher than ETp rates for some periods, the seasonal ETa estimates were lower than the seasonal ETp for all treatments under both nitrogen conditions (Table 7) Seasonal ETa estimates, under nitrogen sufficient conditions were 47, 83, and 70% of the seasonal ETp for treatments 1, 2, and 3, respectively. Percentages of 45, 74, and 68 were estimated for the same water management treatments under nitrogen stress conditions. Water-use efficiency calculations from seasonal water balance data are given in Table 8 Water-use efficiencies represent the percentage of seasonal irrigation amounts that were used to increase ETa. As observed in Table 8 low ETa increases from irrigation were obtained for both treatment 2 and 3 under nitrogen stress conditions. This resulted in a low water-use efficiency in relation to that obtained under nitrogen sufficient conditions. Thus, the well-irrigated treatment 2 under nitrogen sufficient conditions gave the highest water-use efficiency (80%) Water stress in treatment 3 for the same nitrogen sufficient conditions resulted in a 14% reduction in water-use efficiency. In contrast, nitrogen stress in treatments 2 (well-irrigated) and 3 (water-stressed) resulted in similar water-use efficiencies. The above water-use efficiencies are in close agreement with those previously estimated by the ratio of water production functions.

PAGE 82

QJ 60 n) 1—1 i-j CJ > CJ c o uO •H z CO o •rl 4-1 U-l iH OJ 60 CO CN 00 • • • CJ 0) 00 CN UO rH I— 1 rH d) < •H CJ •H 'CJ CN 00 I— t • • • (U O cn CO CM (— 1 I— 1 CO 1 0) c i-i c u E •H c o 4-1 •H n3 CO tC •H O in CO U o • • i-t CN t— i 1— i I— 1 o G c o •H 4.1 vdCO tn CO 00 GO •H a\ CO >^ )J CN CN u M CJ 4-1 CO c
PAGE 83

70 Peanut Experiment Rainfall and Irrigation The rainfall and the detailed irrigation schedule for the growing season of the peanut experiment are presented in Fig. 15 and Table 9, respectively. Total rainfall for the growing season differed between irrigated treatments 2 and 3 because of the sheltering study of treatment 3 (dry cycle) on genotypes UF 781181-1217, PI 383426, Early Bunch, and Florunner. A total of 19.14 cm of rainfall was sheltered from treatment 3 during the sheltering study (from 3 July to 2 August) To establish the crop, all treatments received three small irrigations that totaled 2.69 cm and were recorded as rainfall. Differential irrigation treatments began on day 37 when treatments 2 and 3 were irrigated. Thereafter, irrigation was accomplished for the two irrigated treatments when the required conditions were met. Total amounts of irrigation in these two treatments were similar (Table 9). Rainfall was well distributed throughout the season and rainfall events greater than 2.5 cm were recorded on days 18, 27, 39, 45, 49, 52, 70, 76, 110, and 127. The total amount of rainfall during the season including the initial irrigation of 2.69 cm was 78.83 cm. Yield Response Because the sheltering study in irrigation treatment 3 involved only four peanut genotypes, two statistical analyses were performed on the pod yields. The first one considered treatment 1 (rainfed) and treatment 2 (optimal irrigation) having the 10 peanut genotypes

PAGE 85

72 Table 9. Irrigation schedule, peanuts, 1982, Irrigation amount on treatment number Date Day 2 3 cm June 10 37 1.90 1.90 July 2 59 1.78 1.78 July 9 66 1.68 Aug. 10 98 1.52 2. 54 Sept. 2 121 2.03 2.03 TOTAL 8.91 8.25

PAGE 86

73 as subtreatments The second analysis considered only the four peanut genotypes which were in all three water management treatments. Pod yield (averages of four replicates) for all main plot treatments are summarized in Table 10. As there was no significant interaction between irrigation and genotypes, average yields across genotypes can be used in analysis of irrigation and ETa production functions. The irrigation amount of 8.91 cm did not increase pod yield significantly. These results indicate that seasonal rainfall distribution adequately supplied the peanut evapotranspirative needs. The results are in agreement with the findings of Varnell et al. (1976) in which supplying 10 cm of water in addition to the 70 cm of natural rainfall did not improve the yield of Florunner. As indicated in the footnotes of Table 10, the irrigation regression coefficients were 55 and 57 kg ha cm for 10 and 4 genotype averages, respectively, using treatments 1 and 2. The respective ETa regression coefficients were 159 and 166 kg ha cm Irrigation-use efficiencies were about 34% in both cases. The addition of treatment 3 to the 4 genotype averages provided for a linear regression analysis. The linear function was: Pod yield = 146 + 99 ETa = 0.19, (18) All 48 data points comprising each treatment average were used in the regression analysis. The low coefficient of determination reveals the high variation in the yield data. It is interesting that the ETa regression coefficients, 159 and 166 kg ha cm ^, compare favorably with the 162 value obtained by Hammond et al. (1981b) in 1977 with an irrigation-use efficiency of

PAGE 87

Table 10. Peanut pod yields and simulated evapotranspiration as affected by water management, 1982. Pod yieldsi Treatment Irrigt ETa Ten genotypes^ Four genotypes^ cm kg ha 1 45.49 4866 4534 2 8.91 48.55 5353 5043 3 8.25 37.54 3888 fSeasonal rainfall 78.73 cm for treatments 1 and 2 and 60.24 cm for treatment 3. _2 jFrom treatments 1 and 2, AY/lrrig = 55 and 57 kg ha cm for 10 and 4 genotype averages, respectively. Respective AY/ETa values 159 and 166 kg ha"! cm"!. §Pod yield data courtesy of Drs. K. J. Boote, J. M. Bennett, and L. C. Hammond and Mrs. T. D. Rodriguez.

PAGE 88

75 77%. Moreover, under well-irrigated conditions, the estimated ETa values and pod yields in 1977 were nearly the same as in 1980. Clearly, more irrigation water was applied in the present study than needed during the relatively wet growing season of 1982. Theoretically, the amount of irrigation needed in 1982, assuming a 77% efficiency would be 4 cm. An ETa regression coefficient of 120 kg ha""*" cm""'" was calculated from the data of Pallas et al. (1979). Their study was conducted on Florunner peanuts at Tifton, Georgia. The Ky values of Doorenbos and Kassam (1979) for the 10 and 4 genotype averages were 1.44 and 1.60, respectively. Thus, a 20% ETa deficit would predict respective yield reductions of 29 and 32%. Water Balance Periodic and seasonal water balance data are presented in Tables 11 and 12. Seasonal balances were the summation of periodic water balance values. Measured water depletions were calculated from rainfall and irrigation inputs and neutron readings taken in the Florunner peanut in two replications of each water management treatment. Estimated depletion and its components, ETa and drainage, were obtained by simulation. The data of Table 11 show that there was a close agreement between measured periodic and simulated depletion values throughout the growing season. The largest difference was observed on treatment 3 during the induced stress period (60-83 days) where the simulated depletion was 1 cm lower than measured. After the end of stress, the profile remained at a water content level below the other treatments for about 9 days

PAGE 89

76 Table 11. Periodic water balance during the growth period of ten peanut genotypes, 1982. Measured water Estimated water Treatment Inputf AS| Depletion ETa Drainage AS Depletion§ — cm — 0-16 days (ETp =6.06 cm) 1 2.69 0.37 2.32 2.44 0 0.25 2.44 z 0 ^ Q U 37 2 32 2.44 0 0.25 2.44 3 2.69 0.37 2 32 ? 44 (.Hip — n XX jx cm ) U ZD 9 / /i 1 9.50 -0.26 9.76 4.88 5.39 -0.77 10.27 9 y ju -0 30 9 80 4.93 5. 11 -0.54 10.04 3 9.50 -0.30 9 80 — / w *-rX Ud y o 4 Q"^ /'p'T'-, V,Iiip — J XX 9 A1 r>Tn^ z D X cm } — U Jt xU 1)4 1 5.08 1.50 3.58 1.23 2.26 1.59 3.49 Z b yo 1.53 5.45 1. 92 3.41 1.65 5.33 3 6.98 1.53 5.45 41-48 days 1.92 (ETp = 3.41 3.22 cm) 1.65 5.33 1 8. 15 -0 83 O o -0. 42 8.57 2 8.15 -0.37 8.52 2.74 5.79 -0.38 8.53 3 8. 15 -0. 37 8.52 48-55 days 2.74 (ETp = 5.79 3.00 cm) -0.38 8.53 1 7.13 0.51 6.62 2.64 4.04 0.45 6.68 2 7.13 0.49 6.64 2.64 4.01 0.45 6.68 3 7.13 0.49 6.64 55-60 days 2.64 (ETp = 4.04 2.57 cm) 0.45 6.68 3 1.90 -1.25 3.15 55-70 days 2.52 (ETp = 0.40 7.01 cm) -1.02 2.92 1 2.21 -3.99 6.20 5.09 0.89 -3.77 5.98 2 5.66 -1.91 7.57 60-83 days 6.83 (ETp = 1.06 9.60 cm) -2.23 7.89 3 0 -5.32 5.32 4.29 0.04 -4.33 4.33

PAGE 90

77 Table 11 — continued. Measured water Estimated water Treatment Inputt As| Depletion ETa Drainage aS Depletion^ cm 70-73 days (ETp =1.35 cm) 1 5.69 4.66 1.03 1.17 0.12 4.40 1.29 2 5.69 3.33 2.36 1.41 0.65 3.63 2.06 73-77 days (ETp =1.38 cm) 1 5.33 1.58 3.75 1.51 2.22 1.60 3.73 2 5.33 0.38 4.95 1.51 2.92 0;90 4.43 77-79 days (ETp =0.74 cm) 1 0.17 -1.56 1.73 0.81 1.00 -1.64 1.81 2 0.17 -1.56 1.73 0.81 0.85 -1.49 1.66 79-83 days (ETp = 1.70 cm) 1 2.08 -0.55 2.63 1.87 0.78 -0.57 2.65 2 2.08 -1.08 3.16 1.87 0.78 -0.57 2.65 83-90 days (ETp =2.55 cm) 3 0 -0.61 0.61 0.61 0 -0.61 0.61 83-94 days (ETp =4.25 cm) 1 5.51 -0.14 5.65 4.67 1.01 -0.17 5.68 2 5.51 -0.18 5.69 4.67 1.00 -0.16 5.67 90-97 days (ETp =3.05 cm) 3 2.55 0.98 1.57 1.61 0.04 0.90 1.65 94-100 days (ETp =2.68 cm) 1 0.18 +2.99 3.17 2.95 0.28 -3.05 3.23 2 1.70 -1.20 2.90 2.95 0.02 -1.27 2.97 97-99 days (ETp =0.90 cm) 3 2.54 1.44 1.10 0.76 0.36 1.42 1.12 99120 days (ETp =8.37 cm) 3 11.81 1.59 10.22 7.62 2.76 1.43 10.38 100106 days (ETp =2.44 cm) 1 1.85 -0.07 1.92 2.35 0 -0.50 2.35 2 1.85 -0.61 2.46 2.64 0.01 -0.80 2.65

PAGE 91

78 Table 11 — continued. Measured water Estimated water Treatment Inputt Asi Depletion ETa Drainage AS Depletion^ cm 106-108 days (ETp = 0.67 cm) 1 2. ,03 1. ,35 0.68 0.73 0.10 1. ,20 0. 83 2 2. ,03 1. ,51 0.52 108-116 0.74 days (ETp = 0.01 3.14 cm) 1. ,28 0. 75 1 7. 57 0. 40 7.17 3.29 3.70 0. 58 6. 99 2 7. 57 -0. 22 7.79 116-131 3.29 days (ETp = 4.24 5.05 cm) 0. 04 7. 53 1 10. 82 2. 12 8.70 4.89 3.86 2. 07 8. 75 2 12. 85 2. 06 10.79 120-139 4.91 days (ETp = 5.92 5.85 cm) 2. 02 10. 83 3 15. 24 2. 38 12.86 131-139 5.46 days (ETp = 7,44 2.50 cm) 2. 34 12. 90 1 2. 74 -0. 88 3.62 2.25 1.47 -0. 98 3. 72 2 2. 74 -1. 27 4.01 2.25 1.49 -1. 00 3. 74 fRainfall plus irrigation. $Soil profile water content on last day of period minus content on first day. §Estimated ET plus estimated drainage.

PAGE 92

M-i O u 1 1 r\ i> rrt Tj ni 00 CO I 1 lO CO < e o rH rH •H 4J tn c O CM *H oi i_i rH CO CO (Tl LO lO LO 00 NM e rH CJ CO CO 111 w 'H rH 1 1 lO LO V-l CL) QJ 1 o 00 dl U.f ^ f— 1 r* ni o > C/^ CN CO y\ CNj HI ,__| r* CO OJ Cu (- w Tt en CO o> ni rH a1 *y rrt oo 00 > 1 J vO rH cd O CO CO O CO rC O CO CO 0\ 0^ iH GO CO CO CO CO CO fTl
PAGE 93

80 (90-99 days). In general, there was a consistent equality of neutronbased and simulated AS values which contrasted with the periodic divergent data obtained in the corn experiment because of a malfunctioning neutron-meter. For the peanut experiment, the seasonal difference between measured and simulated depletion was less than 1 cm (Table 12) Figure 16 also shows a close agreement between measured and simulated root zone water contents for all treatments. Note that drought from rain-sheltering treatment 3, was severe enough to reduce the root zone water content to less than 15 bars content. Figure 16 also reflects the good rainfall distribution during the growing season (treatment 1). Thus, there was no marked difference in profile water depth between the rainfed and well-irrigated treatments. Note from Table 12 that 8.9 cm of irrigation on treatment 2 increased ETa by 3.06 cm. So a low water-use efficiency of 34% (3.06/8.9 = 0.34) was obtained for the wet peanut growing season. Simulated ETa values were lower than ETp (Penman, 1948) from planting to day 55 for all treatments. During mid-season (73-116 days) simulated ETa exceeded ETp in treatments 1 and 2. For the stressed treatment, it is interesting to note that in the 55-60 days period preceding the induced water stress, simulated ETa had almost reached estimated ETp. During the induced stress simulated ETa was reduced significantly. After the relief of stress (120-139 days) ETa values were increased to almost equal ETp. Seasonal ETa estimates were about 77, 82, and 64% of the seasonal ETp for treatments 1, 2, and 3, respectively.

PAGE 94

16. Measured and simulated root zone water content distributions with time under rainfed (treatment 1), well irrigated (treatment 2) and irrigated but induced dry cycle (treatment 3) conditions, peanut genotypes.

PAGE 95

82 Soybean Experiment Rainfall and Irrigation Rainfall and irrigation distribution for the growing season of 'Cobb' soybeans are shown in Fig. 17 and Table 13. All treatments received three initial irrigations on days 2, 4, and 9 (total = 2.34 cm) and they were recorded as rainfall. There were four drought periods in which irrigation was used. Differential irrigation treatments began on day 40 when treatments 4, 5, and 6 were irrigated. Thereafter irrigations were accomplished for all irrigated treatments when the required conditions were met. Figure 17 and Table 13 show that irrigations were applied in small amounts (less than 2,0 cm) to partially refill the depleted rooting zone. Seasonal rainfall varied with treatment (see Table 13). Yield Response The linear regression analysis between soybean yield and seasonal irrigation is presented in Fig. 18. The rainfed treatment gave unusually high yields when compared to the linear response to irrigation exhibited by the irrigated treatments. It is the opinion of Drs. K. J. Boote and L. C. Hammond (personal communication) that the rainfed yields are biased upward because of yield sampling to avoid the spatial variability present in the experimental plots. Including the rainfed plots in the analysis resulted in the following response function: Y = 2265 + 39 (IRRIG) R^ = 0.59.

PAGE 96

< H H < H H Z < OS EZ S H < H Z a <: ca — O O — I O (ici) Noiivai^i H CO O H CO > o sr as s: Cd > o z u o Eo CM CO 0^ i-I o z Z Ed O 2 Z Ed M H Z Cd < W o a vO E-> b < CO E-1 C > CO in < 3 Q O < J 3: O I o c o cn tn tc c •H o u 0) C •H i-' D V C o •H ca 6C •H -l ^1 T3 C M-l c •ri 03 cw O C O CO nj t— I c cn CO C to c
PAGE 97

84 Table 13, Irrigation schedule, 'Cobb' soybean, 1982. Irrigation amount on treatment numbers Date Dayt 2 3 4 5 6 cm Aug. 9 40 1.52 1. 52 1 CO 1.52 13 44 1 CO 1 5o T CO l.o8 1 CO 1. 5o 27 58 1. 52 1. 52 1 CO 1. 52 T CO 1.52 31 62 1.52 1.52 0. 84 1 CO 1.52 Sept. 2 64 1 52 1. 52 o r\ o 2 03 T CO 1.52 16 78 1.78 1.78 1.78 1.78 Oct. 8 100 1.52 1.52 1.78 12 104 1.90 1.90 1.90 18 110 1.27 1.27 20 112 1.78 22 114 1.27 1.27 TOTAL 6.34 12.30 10.08 7.49 15.66 fSeason length treatments 3, : 118 days for treatments 1, 2, and 5; 127 days 4, and 6. Respective seasonal rainfalls were 5 for 8.04 and 54.20 cm.

PAGE 98

85 o 6 a 2 O M H <: o M en M o CO -) C • O e • Cfl 03 U r-H Q C3 (^01 X Tc o o CO cfl >^ OJ 03 03 OJ O >-( 4-1 3 o 03 a c (fl OJ o iH 03 rsl x: XI XI 3 o o u x: 4J y-l o 00 0) 03 C 03 ^ O (X •H C 03 O 01 ^-1 ij-i P O cfl t3 a: rH cfl OJ u • •H cfl CJ c T3 ^J •H iH tfl H C o >^ Cfl 00 •H

PAGE 99

86 Hammond et al. (1981b) obtained an irrigation response coefficient of 84 kg ha cm for Bragg and Cobb soybeans in 1978. Figure 19 shows results of linear regression analyses between soybean grain yields and simulated seasonal evapotranspiration. Two linear relationships were well-defined. The first one was obtained from treatments 6 (well-irrigated), 3, and 2. In these last two treatments, water stress occurred during vegetative growth. The second relationship involved treatments 6, 4, and 5. In treatments 4 and 5, water stress occurred during pod set and pod filling stages. As observed in Fig. 19, the rainfed treatment was not included in the regression analysis. However, when included with treatments 2, 3, and 6, the coefficient was decreased to 63 kg ha cm but the determination coefficient was reduced from 0.91 to 0.84. With all treatments included, the function was: Y = 754 + 59 (ETa) R^= 0.65. The determination coefficients tend to justify the selection of the two functional relationships in Fig. 19. The existence of two ET production functions in the present study is in line with the argument of Stewart et al. (1976) favoring the potential for a family of curves depending upon sequencing of ET deficits during the growing season. Others have emphasized the interaction of water stress and growth stages (Denmead and Shaw, 1960) Severe yield reduction as a result of water stress during pod set and pod filling stages of soybeans have been found by several researchers (Doss et al. 1974; Sionit and Kramer, 1977; Smajstrla and Clark, 1981; Ambak, 1982).

PAGE 101

88 Both the irrigation and ETa production coefficients have water management implications. The irrigation production coefficient is smaller than the ETa coefficient (Stewart, 1976; Skogerboe et al. 1979; Stegman et al. 1980; Hammond, 1981b). The goal of efficient use of irrigation is achieved when all water supplied by irrigation meets the crop ET needs imposed by the atmosphere. In humid regions like Florida, it is not possible to avoid non-ETa losses because of rainfall uncertainty. The losses can be minimized through the use of judicious irrigation scheduling strategies and practices. The development and testing of alternative strategies is aided by evaluation of production functions. When irrigation and ETa functions are linear, the ratios of their coefficients provide a measure of the overall efficiency of the several levels of irrigation used in the particular season. Using the irrigation coefficients from Fig. 18, we calculated two irrigation-use efficiencies for the two sets of treatments in Fig. 19. The efficiencies were 76/75 = 1.01 and 76/132 = 0.58 for water stress during early vegetative and pod set-pod filling periods, respectively. The regression ratio with all treatments included was 39/59 = 0.66. One would not expect the 100% efficiency value obtained for treatments 2, 3, and 6. This result suggests that the selection of two ETa production functions from these data may not be justified in contrast to the earlier justification on the basis of coefficients of determination. In Fig. 20 are plotted the relative soybean grain yields (Ya/Ym) as a function of relative seasonal evapotranspiration (ETa/ETm) Stegman et al. (1980) point out that intercepts of the regression lines with the ETa/ETm axis tend to be low when crops are stress tolerant or when stresses are ideally distributed between the vegetative, reproductive

PAGE 102

1.0-, 0-0 0.2 0.4 0.6 0.8 ETa/ETm Fig. 20. Relative grain yield (ya/Ym) of 'Cobb; soybeans versus relative seasonal evapotransoiration (ETa/ ETm)

PAGE 103

90 and grain filling periods of growth. Intercepts at high ETa/ETm end are likely when stresses are extreme and in the most sensitive growth stages of stress sensitive crops. The area between the higher and lower curves indicates the relative yield variation for a given relative seasonal ET, depending on the timing of stresses. Thus, for example, 80% of the seasonal relative ET will cause 77 to 64% of relative grain yield depending on the water management conditions during the growing season. A similar analysis in terms of percentage reduction in yield and percentage seasonal ET deficit was used by Doorenbos and Kassam (1979). The yield-ETa data plotted in this way (Fig. 21) produces the same graphic display as in Fig, 20. The coordinates differ, but the Ky values are the same. Crop response factors (Ky) close to or less than 1 indicate that yields were proportionally reduced by water stress. Conversely, Ky values greater than 1 indicate that grain yields were strongly reduced by water stress. Thus, the larger the Ky value the larger the yield decrease per unit decrease in ETa. From the present analysis, it can be established that a 20% relative ETa deficit during vegetative growth would produce a 18% relative yield reduction (0.9 x 0.20 = 0.18). However, with the same relative ETa deficit during reproductive and pod filling periods there would be a 31% reduction in yield (1.57 x 0.20 = 0.31). The above analysis of the yield response to water deficit in different growth periods is of major importance in the scheduling of available but limited water supplies for highest yields. This implies that timing of irrigation is as crucial as the total amounts applied

PAGE 104

91 Fig. 21. Relationship between relative grain yield decrease (1 Ya/Ym) and relative seasonal evapotranspiration deficit (1 Eta/ETm) for 'Cobb' soybeans, 1982.

PAGE 105

92 over the growing season. Thus, in terms of water management strategy Doorenbos and Kassam (1979) point out that water allocations of a controlled but limited supply would be directed toward meeting the full water requirements of the crop during the most sensitive growth stages rather than spreading the available but limited supply to the crop equally over the total growing period. For example, for soybean production the supply would be directed particularly to the reproductive and pod filling stages. In humid areas where water resources are not limited, maximum return from irrigation requires avoidance of ET deficits throughout the growing season. In Florida, because welldrained sandy soils store temporarily less than 2,5 cm per 30 cm of soil depth, short periods of drought often occur between periods of heavy precipitation (Hammond et al. 1981b). Hence, irrigation in Florida is increasing in importance, especially with high-value crops. Figure 22 summarizes the basic field relationship between percentage reduction in soybean production and percentage seasonal ET deficit from research findings obtained in the IREP by several research workers (Hammond et al. 1981b; Ambak, 1982) as well as the result of the present water management experiment. The purpose of this relationship was to define an overall crop response factor for soybean production under Florida conditions. The scatter of the data in Fig. 22 can be attributed to differences in ETa estimates (based on Penman and daily water balance computations), differences in soybean cultivars, year variations, and timing of ET deficits. The effects of water stress and reduced evapotranspiration at various growth stages in grain production of soybean is emphasized in Fig. 22.

PAGE 106

. 22. Relationship between relative yield reduction (1 Ya/Ym) and relative seasonal evapotranspiration deficit (1 ETa/ETm) for soybean production at Gainesville, Florida.

PAGE 107

94 Water Balance Periodic and seasonal measured and simulated water balance for nondef oliated 'Cobb' soybeans are presented in Tables 14 and 15. The total seasonal water balance was calculated as the summation of measured and simulated values throughout the growing season. Measured water depletions were calculated from rain and irrigation inputs and neutron measured soil water content. The latter were obtained at single sites in each water management treatment in replication 2. Estimates of ETa, drainage and profile depletions were obtained from computer simulation (Rao at al. 1976, 1981). The drainage loss of 16 cm during the period of 0 to 37 days represented about 45% of the total drainage loss for the season. In treatments 2 and 3, the seasonal drainage losses were up to 57% before irrigation was used. In general, there was a close agreement between measured and simulated periodic and seasonal water depletion values. Lack of agreement in treatments 2, 3, 5, and 6 was observed during periods 58-64 and 64-71 days. Simulated depletions were lower than the measured values for the first period, but the contrary was true in the next period. It seems that the measured water content on day 64 was too low relative to the measured values on day 58 and 71 (Appendix Table 29). If we combine period 58-64 and 64-71 days into one (58-71 days), then we find for treatment 2 that measured and simulated depletions are 8.84 and 9.54 cm, respectively. Now, if we make the new period even larger by going back to day 55 (55-71 days), then the measured and simulated depletions are 11.74 and 11,80 cm, respectively. The difference was only 0.06 cm.

PAGE 108

95 Table 14. Periodic water balance during the growth period of nondefoliated 'Cobb' soybeans, 1982. Measured water Estimated water Treatmentf InputJ aS§ Depletion ETa Drainage AS§ Depletion cm 0-12 days (ETp =5.47 cm) 1-6 4.42 -0.06 4.48 2.45 1.92 0.05 4.37 12-29 days (ETp =6.48 cm) 1 15.32 0.21 15.11 4.01 11.10 0.21 15.11 2 15.32 0.21 15.11 4.01 11.10 0.21 15.11 3 15.32 0.21 15,41 4.01 11.10 0.21 15.11 4 15.32 0.21 15.11 4.01 11.10 0.21 15.11 5 15.32 0.02 15.30 4.01 11.10 0.21 15.11 6 15.32 0.02 15.30 4.01 11.10 0.21 15.11 29-37 days (ETp = 3.43 cm) • 1 3.71 -0.23 3.94 2,57 1.74 -0.60 4.31 2 3.71 -0.17 3.88 2.57 1.74 -0.60 4.31 3 3.71 -0.13 3.84 2.57 1.74 -0.60 4.31 4 3.71 -0.61 4.32 2.57 1.74 -0.60 4.31 5 3.71 -0.01 3.70 2.57 1.74 -0.60 4.31 6 3.71 -0.59 4.30 2.57 1.75 -0.60 4,31 37-48 days (ETp =4.80 cm) 1 1.70 -1.28 2.98 2.00 0.54 -0.84 2.54 2 1.70 -1.42 3.12 2.00 0.54 -0.84 2.54 3 1.70 -0.67 3.37 2.00 0.54 -0.84 2.54 4 4.80 -0.16 4.96 4.15 0.70 -0.05 4.85 5 4.80 0.13 4.67 4.15 0.70 -0.05 4.85 6 4.80 0.07 4.73 4,15 0.70 -0.05 4.85 48-51 days (ETp =1.00 cm) 1 2.36 1,39 0.97 0.82 0.16 1.38 0.98 2 2.36 1.99 0,37 0.82 0.16 1.38 0.98 3 2.36 1.60 0.76 0.82 0.16 1.38 0.98 4 2.36 1.09 0.97 0.93 0.74 0.69 1.67 5 2.36 1.66 0.70 0.93 0.74 0.69 1,67 6 2.36 1.59 0.77 0.93 0.74 0.69 1,67

PAGE 109

96 Table 14 — continued. Measured water Estimated water Treatmentt Input? AS§ Depletion ETa Drainage aS§ Depletion cm 51-55 days (ETp = 1.48 cm) 1 7.21 1.06 6.15 1.27 5.49 0.45 6.76 2 7.21 0.83 6.38 1.30 5.48 0.43 6.78 3 7.21 1.17 6.04 1.30 5.47 0.44 6.77 4 7.21 0.29 6.92 1.39 5.55 0.27 6.94 5 7.21 -0.08 7.29 1.49 5.53 0.19 7.02 6 7.21 -0.27 7.48 1.51 5.52 0.18 7.03 55-58 days (ETp =1.24 cm) 1 0.36 -2.08 2.44 1.07 0.89 -1.60 1.96 2 1.88 -0.55 2.43 1.09 0.69 0.10 1.78 3 1.88 -0.55 2.43 1.09 0.69 0.10 1.78 4 0.36 -1.25 1.61 1.17 0.84 -1.65 2.01 5 1.88 -0.71 2.59 1.26 0.66 -0.04 1.92 6 1.88 -0.73 2.61 1.29 0.78 -0.09 2.08 58-64 days (ETp = 2.47 cm) 1 0.36 -1.73 2.09 1.67 0.38 -1.69 2.05 2 3.40 -1.93 5.33 2.17 0.61 0.62 2.78 3 3.40 -1.75 5.15 2.17 0.60 0.63 2.77 4 0.36 -1.46 1.82 1.72 0.36 -1.72 2.08 5 3.23 -2.28 5.51 2.52 0.00 0.00 2.52 6 3.40 -1.70 5.10 2.57 0.60 0.23 3.17 64-71 days (ETp =2.16 cm) 1 7.29 4.68 2.61 1.34 1.32 4.63 2.66 2 7.29 3.87 3.43 1.97 4.73 0.59 6.70 3 7.29 3.68 3.61 1.97 5.21 0.11 7.18 4 7.29 4.87 2.42 1.39 1.19 4.71 2.58 5 7.29 3.46 3.83 2.06 3.69 1.54 5.75 6 7.29 1.90 5.39 2.35 3.64 1.30 5.99

PAGE 110

97 Table 1^ — continued. Measured water Estimated water Treatment! Inpu Depletion ETa Drainage AS§ Depletion 71-77 days cm (ETp = 1.83 cm) 1 i. bl -2.67 6 28 1.61 5 .07 -3.07 6.68 7 97 1 68 -9 fil ^ ox ft A9 J J. 61 -4. 28 7 89 4 on H • \J\J -9 07 S ftR _J o o J ol -3 5? 1 7ft L 79 -9 S7 ft ZtS 0 HO 5 3.61 -3. 69 7 30 1 96 H• D U -9 QS ^ yj ft Sft 0.30 6 3.61 -3.65 7 26 77-86 days 9 09 (ETp = 2.48 cm) -9 8A ^ OH ft AS D H J 1 T p J JO -U. 5Z 4 11 2 23 2. 20 -0.85 4.43 z D JO 1.06 4 30 9 3^^ 9 fi4 OH u • jy A 07 H y / D Jo -1. 20 6 56 9 SI 9 S4 u ox J 3d 0.55 4 81 9 SI ^ 1 7 -0 ^^9 S ftS J Do 5 3.58 0.05 3 S3 9 91 -n 70 H.J/ 0 J JO 0. 81 4 55 86-93 days ? 7'^ Z. • / J (ETp = 9 QQ 1.87 cm) — U JO 79 J / Z 1 1 3 56 1. 72 1.84 1 63 1 67 X Q / n 9ft I 3.56 0.04 3 S? 1 79 "^Q J J7 — X J J J XX J 3.56 0 1 S -3 H X • 7X J y J 9 9 8 — Z Zo D • OH 4 3. 56 9 77 X y X 1 9 / J Z4 -1.59 5 15 5 3.56 X J7 X JD U 41 J 15 6 3.56 0 AO 1 J • xo 93-99 days 9 no (ETp = J Uo 1.74 cm) -1. 54 5 .10 1 1.30 -2.15 3.45 1.38 1.77 -1.85 3.15 2 1.30 -2.20 3.59 1.48 1.90 -2.08 3.38 3 1.30 -2.21 3.51 1.76 1.67 -2.13 3.43 1.30 -2.87 4.17 1.77 1.62 -2.09 3.39 5 1.30 -3.13 4.43 1.45 1.78 -1.93 3.23 6 1.30 -2.06 3.36 1.76 1.26 -1.72 3.02

PAGE 111

98 Table 14 — continued. Measured water Estimated water Treatmentt Input? Asi Depletion ETa Drainage asI Depletion cm 99-106 days (ETp =2.04 cm) 1 0.00 -2.89 2.89 1.30 0.42 -1.72 1.72 2 0.00 -3.13 3.13 1.35 0.40 -1.75 1.75 3 3.43 0.37 3.06 2.08 0.53 0.82 2.61 4 3.43 0.12 3.31 2.07 0.46 0.90 2.53 5 0.00 -2.60 2.60 1.35 0.41 -1.76 1.76 6 3.43 0.69 2.74 2.07 0.55 0.81 2.62 106-113 days (ETp = 1.87 cm) 1 0.00 -1.14 1.14 0.71 0.04 -0.75 0.75 2 0.00 -1.03 1.03 0.76 0.02 -0.80 0.80 3 1.27 -1.30 2.57 1.73 0.11 -0.43 1.70 4 1.78 -0.45 2.23 1.65 0.10 0.31 1.47 5 0.00 -1.00 1.00 0.74 0.04 -0.78 0.78 6 1.27 -0.71 1.98 1.80 0.13 -0.50 1.77 113-119 days (ETp = 1.52 cm) 1 7 Q i • / O 1 11 0.67 0 38 0. 10 1. 30 0.48 2 1.78 1.43 0.35 0.53 0.09 1.16 0.62 3 3.05 2.36 0.69 1.24 0.22 1.59 1.46 4 1.78 0.40 1.38 1.17 0.47 0.14 1.64 5 1.78 1.73 0.05 0.40 0.11 1.27 0.51 6 3.05 1.66 1.39 1.32 0.27 1.46 1.59 119-127 days (ETp = 1.95 cm) 3 3.84 1.70 2.14 1.17 1.62 1.05 2.79 4 3.84 2.72 1.12 1.12 1.17 1.55 2.29 6 3.84 1.56 2.28 1.27 1.24 1.33 2.51 fSee p. 27 for description of treatments. ^Rainfall plus irrigation. §Measured and estimated soil profile water content at the first and last day of each time period.

PAGE 112

u JJ cd 0) s •H 4J 03 W H 4-1 3 CO cS s Q d o •H 4-1 OJ rH a Q cos 4-) 3 a o l4-i o ++ c j:: o cn 00 CO C N 00 CO iH X) rH 00 r-. 00 T-l CN CN CN .H rH rH CO in to 4J O c rH 6 CO 4-J 3 CO CO OJ •H U > 4-1 C UH 0 o 13 c -i to •H Xf CO CO 4-1 3 O 00 rH iw c a. CN iH C CO a O 14-1 to c OJ CO -H 0) (U CO tz) CO Di + — Hcoo

PAGE 113

100 There are other factors which could be involved in these descrepancies of periodic depletion values. The adjustment in rooting depth requires the addition of a new layer of root-free soil to the initial profile of the period. With the added uncertainty of root depth and activity, one would expect less precise water depth estimates for this layer over a wide range of water flow conditions. Another factor is the fact that the daily simulated results are based on a 24-hour time frame, while the measured water contents are based on a 5 to 15-min time frame. A rainfall and/or irrigation event is input to the model at the beginning of the day regardless of when the input occurred during the day. Thus, the simulated water redistribution condition (Day-end) could be quite different than the measured condition at specific times earlier in the day. Measured and simulated water content distributions with time for treatment 1 (rainfed) and treatment 6 (well-irrigated) are shown in Fig. 23. In both treatments, close agreement was observed during the period of 0 to 86 days. From day 86 to the end of the season, the simulated water depths (root zone) were slightly lower than the measured values. It appears that the model is not precisely calibrated in terms of input parameters such as, field capacity, root depth with time, residence time for water redistribution, and crop coefficients. The AS values in Table 14 reflect the differences in water contents between the associated successive points in Fig. 23 only after maximum root depth had been attained.

PAGE 114

Measured and simulated root zone water content dist bution with time for treatment 1 (rainfed) and trea 6 (well irrigated), 'Cobb' soybeans, 1982.

PAGE 115

102 Comparisons between measured and simulated water content distributions for treatments 2, 3, 4, and 5 are given in Appendix Table 29. There was close agreement between measured and simulated values in all treatments throughout the growing season. The above data indicate that the current simulation model provided good estimates of water content in the root zone. Thus, the simulation model could be used for scheduling irrigation. Irrigations would be triggered when the predicted water depth in the root zone reached some preassigned level. The well-irrigated treatment 6 gave periodic ETa values greater than ETp from periods 51-55 to 99-106 days as a result of Kc values greater than 1. The ETa rates of treatment 5 also exceeded the ETp • rates during mid-season (51-55 to 71-77 days period) but later in the season the ETa rates decreased to reach the ETa rates of the rainfed treatment. On the rainfed treatment, because Kc values were lower than 1 throughout the se ason, the ETa rates never reached the ETp rates. Irrigation-use efficiencies for individual treatments are given in Table 16. The lowest efficiency of all treatments was given by treatment 2, while treatment 4 gave the highest. Average irrigationuse efficiencies given in Table 16, are in close agreement with those estimated by the ratio of the two production coefficients when irrigation was withheld in pod set and pod filling periods (58%)

PAGE 116

103 Table 16. Irrigation-use efficiency for 'Cobb' soyb eans, 1982. Treatment Irrigation ETa AETaf Efficiency cm 1 26 4 2 6.3 28.2 1.8 0.29 3 12.3 32.5 6.1 0.50 10.1 33.7 7 3 0. 72 5 7.5 30.9 4.5 0.60 6 15.7 36.8 10.4 0.66 AVERAGE 10.38 31.42 6.02 0.56 fEvapotranspiration increase from irrigation.

PAGE 117

104 Sweet Potato Experiment Rainfall and Irrigation The rainfall and irrigation distribution for the growing season of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes are shown in Fig. 24 and Table 17. Rainfall events greater than 4 cm were recorded on days 15, 23, and 63. Four significant drought periods occurred: 30-40, 50-60, 90-105, and 120-135 days. Differential irrigation treatments began on day 33 when treatments 2 and 5 were irrigated. Thereafter, irrigations were accomplished for all irrigated treatments when the conditions were met. Water stress in treatments 5 and 6 was imposed during the periods 40 to 72 and 72 to 100 days, respectively. Seasonal rainfall of 56.70 cm was recorded on treatments 1, 2, 3, and 4. Rainfall amounts of 24.36 and 10.21 cm were kept off treatments 5 and 6, respectively, during the stress periods. Yield Response Potato fresh and dry weights, vine dry weights, and total dry weights and irrigation amounts for 'Georgia Jet' and 'Yellow Jewel' are presented in Table 18. In the case of sweet potatoes, the tuberous root fresh weight is the marketable yield. There was a variety by irrigation interaction in potato fresh weight as well as potato dry weight. Note that these two parameters responded in a parallel way. For vine dry weight and total dry weight there was no water management by variety interaction. In all the above responses water

PAGE 118

105 < < oi 2 < z < til E-=7 < CO o CM > o z CO •< X o o K ca > o o w c to ^ o o o o o u z M O H CO z < Ed w r H H < dz a =3 (HO) Moiivoiirai -cCN o (KD) TIVINIVH o I o CO dJ cn &c (1; 1— 4J oc u 3 n Q CO *H J-l 1— 1 tti 60 Q U 4J cd 4J Q T) c n3 U 0) — [ cn — 1 QJ >> QJ Q Q t—l ,—1 o 0) 1^ ca T) "13 C C ca cn c o •H o U 3 ta •H •H >J 60 u cn o •H di CJ QJ x: H o CN 6C •H

PAGE 119

106 Table 17. Irrigation schedule, 'Georgia Jet' and 'Yellow Jewel' sweet potato, 1982. Irrigation amounts on treatment number Date Day 2 3 4 5 6 cm Aug. 10 33 1.82 1.82 11 34 1.82 12 35 3.71 13 36 1.82 1.82 16 36 1.82 1.82 Sept. 1 55 1.82 i.82 4 58 1.82 18 72 1.82 Oct. 9 93 1.82 10 94 1.82 18 102 1.82 20 104 1.82 21 105 1.82 TOTAL 10.92 7.35 3.64 7.28 3.64

PAGE 120

tn o 4J 0) CO a o p< CO (U u c •H 0} M 2 > 4J iH 3 U O CO u o a 4J CO <^ H 0) (U 13 O H H 13 C n) ^— \ • (N 1-3 00 0 ^ — ^ iH — •* 4-1 0) cn I-) u c cfl •H e bD U to 0 OJ c 0 0 •H J-l to to iH •H 0) •H •H to X 0) tu u c to -H ns u -a jj to -H 4-1 tU O S H u u 601 tU -H c w +(3 n) 4) H CO O CO X 60 00 m in CT\ 0 in 0 in 0 0 0 0 0 0 0 00 . CO ^ — V V — ^ CN CO CN CN in cn CO 14H o ti o •H 4J & •H !-< O tn o )-i H O W 14H a) • CN 4J a) CO 60 a 6 CO •H ^ ID 4J 0) oj tn > w w < ++C03 I •H CO tn 4-1 C 0 0 c •H 4-1 (U to -l to •H rH a i a to rH c a 0 •H (U 4J e to to 00 CO •H !-i OJ X •H 4-J 14H c 0 •H CO M ID QJ 4-1 4-1 -i 0) 3 rH C (U CU 6 S-i CO to CO to (U QJ X to 4J OJ XI >^ 4-1 X c (U •U M (U to 3 CU 0 rH c rH •H u
PAGE 121

108 management treatments and potato cultivars were significant (0.05 level). For 'Georgia Jet' treatment 2 potato yields were significantly higher than in treatments 6, 5, 4, and 1, but not in treatment 3. Treatment 1 (rainfed) gave significantly lower yields than treatments 2, 3, 6, and 5, but not treatment 4. For 'Yellow Jewel', potato yields in treatment 4 were higher than in treatments 5, 1, and 6, but not in treatments 2 and 3. There were no significant differences among treatments 2, 3, 5, 1, and 6. Note that 'Georgia Jet' responded differently to treatment 4; yields were not different from treatment 1 (rainfed) The marked difference in cultivar response is seen also in the potato yields in treatments 5 and 6. The apparent stimulating effect of drought stress was found also by Sajjapongse and Roan (1982), where yield increases were associated with increases in tuberous root formation. On the other hand, a lack of response to water stress ('Yellow Jewel' cultivar) could be related to the extensive root system typical of sweet potatoes (Vittum and Flocker, 1967). Jones (1961) found no significant increase in total yield of U.S. No. 1 sweet potatoes when the soil was allowed to dry to only 20% of the available soil water at the 30 cm depth. Averaged over two sweet potato cultivars, vine dry weights of treatments 1, 2, and 4 were significantly higher than the stressed treatment 6, but they were not significantly different than treatments 3 and 5. The yields of these two treatments were not significantly different than treatment 6. The average vine dry weight of all water management treatments on 'Georgia Jef cultivar was 9% lower than the average vine dry weight of 'Yellow Jewel' sweet potato. Total

PAGE 122

109 dry weight of treatments 2 and 3 were significantly higher than treatments 5, 6, and 1, but they were not significantly different than treatment 4. In contrast with the oveall average vine dry weight, the 'Yellow Jewel' cultivar gave 10% less total dry weight than the 'Georgia Jet' variety. In all water management treatments, 'Georgia Jet' gave a higher harvest index than 'Yellow Jewel'. Yield responses of both 'Georgia Jet' and 'Yellow Jewel' cultivars to seasonal irrigation amounts are presented in Fig. 23. Treatments 5 and 6 were not included in the regression analyses because the rainfall base was less than for the rainsheltered treatments. There was a well defined linear relationship between irrigation and yields of 'Georgia Jet'. In contrast, for 'Yellow Jewel' total dry weight only, there were significant irrigation and ETa regressions. Linear regressions between sweet potato yield and seasonal evapotranspiration are presented in Fig. 26. Regression analyses of 'Georgia Jet' data were based on treatments 1, 2, 3, and 4 only. Treatments 5 and 6 were not included because of their large deviation from the linear response exhibited by the other four treatments. The estimated ETa for stressed treatments 5 and 6 could be lower than actual due to unaccounted for surface and subsurface flow of rainfall under the shelters. Rainshelters were not totally efficient during the stress periods. In addition, water extraction from soil depths greater than assumed in the simulation could have occurred; plants of both cultivars did not show severe wilt during the stress periods. On the other hand, the discontinuity exhibited in the 'Georgia Jet' response was not present in the 'Yellow Jewel' results which showed little response to irrigation.

PAGE 123

110 H X o M a o Se X cn bl a: p b OS 00 o EX m U o 14 12 • o 3 Y = 5064 + 406 (IRRIG) = 0.97 Y = 10526 + 377 (IRRIG) = 0.981^ 10 O 2 ^ ^ 10s^ as I a re ^ ^ 8H < 00 rJ! O ^ ^ 611478 + 93 (IRRIG) R^ = 0.44 • GEORGIA JET O YELLOW JEWEL 10 SEASONAL IRRIGATION (CM) Fig. 25. Yields of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes versus seasonal irrigation (IRRIG) amounts, 1982.

PAGE 124

Ill I i 1 I r 24 28 32 36 40 6 • GEORGIA JET O YELLOW JEWEL 0 \ f \ — 1 r 24 28 32 36 40 SEASONAL EVAPOTRANSPIRATION (CM) Fig. 26. Yields (Y) of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes versus seasonal ETa, 1982.

PAGE 125

112 The total dry weight-ET coefficient of 'Yellow Jewel' (221 kg ha""*" cm ^) was 74% lower than that of 'Georgia Jet'. As indicated in previous experiments, irrigation-use efficiency can be measured as the ratio of the irrigation/ET production coefficients. Thus, for marketable yield, 'Georgia Jet' irrigation-use efficiency was 2348/5188 = 45%. This indicates that about 55% of the total water applied by irrigation was allocated as non-ETa losses Doorenbos and Kassam (1979) suggested the method of data presentation in Fig. 27. 'Georgia Jet' data show the relationship between the relative marketable yield decrease and seasonal relative ET deficit. The Ky value of 3.52 indicates the high yield response to evapo transpiration. It is observed that the rainfed treatment underwent a relative ETa deficit of 13% causing a relative yield decrease of 46%. Root Length Density Distribution Non-tuberous roots of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes were sampled at harvest in treatments 1 (rainfed) and 2 (optimal irrigation) The samples were taken in single sites of replications I, II, and III. The results of an analysis of variance (not shown) indicated that there were no significant differences (0.05 level) in irrigation and variety. However, root length density distribution with depth was highly significant. The root sampling was too limited in number and dates to detect treatment effects.

PAGE 126

Fig, 27. Relationship between the relative sweet potatoes fresh weight yield decrease and the seasonal relative ETa deficit for the 'Georgia Jet' variety.

PAGE 127

114 Overall averages of root length density distribution with depth are shown in Fig. 28. The distribution pattern was close to an exponential function. Roots extended to a depth of at least 240 cm with 39, 63, 76, 87, and 93% of the roots occurring in 15, 30, 60, 90, and 120 cm soil depths, respectively. Jones (1961) investigated the root fresh weight distribution of 'Georgia Red' cultivar in a 120 cm profile of Norfolk sandy loam soil. He found 51 and 81% of the roots in the 0-23 and 0-45 cm layers, respectively. In the present study the rooting depth was at least double the maximum rooting depth reported by Jones. Water Balance Periodic and seasonal measured and simulated water balance computations for 'Georgia Jet' sweet potatoes are presented in Tables 19 and 20. Neutron soil water measurements were not made in 'Yellow Jewel' subplots. Total seasonal water balance was calculated as the summation of the periodic water balances. Measured water depletions were calculated from neutron readings, root depths (Appendix Figure 30) and irrigation and rainfall inputs (Table 17 and Fig. 24). Neutron data were taken in treatments 1 and 2 of replication III, treatments 3 and 4 of replication IV, and treatments 5 and 6 of replication II. Estimated ETa, drainage and soil water depletion were obtained from the simulation results. As shown in Table 19, large drainage losses occurred in periods 0-14 and 51-71 days, accounting for about 50% of the total seasonal drainage. In general, results of the periodic and seasonal water

PAGE 128

g. 28. Average root length density distribution of 'Georgia Jet' and 'Yellow Jewel' sweet potatoes at harvest time 1982.

PAGE 129

116 Table 19. Periodic water balance during the growth period of 'Georgia Jet' sweet potato, 1982. Measured water Estimated water Treatmentf Inputj: aS§ Depletion ETa Drainage aS§ Depletion^ 0-14 days cm (ETp = 5 .26 cm) 1-6 12.27 0.31 11.96 14-24-days 2.40 (ETp = 3 9.36 .97 cm) 0.51 11.76 1-6 3.90 -1.17 5.07 24-33 days 2.33 (ETp = 3 2.89 .72 cm) -1.32 5.22 1 3.85 -0.78 4.63 2.96 1.99 -1.10 4.95 2 5.67 0.54 5.13 3.34 1.99 +0.34 5.33 3 3.85 -0.78 4.63 2.99 2.16 -1.30 5.15 4 3.85 -0.78 4.63 3.29 1.86 -1.30 5.15 5 5.67 0.69 4.98 3.34 1.99 n 3A D J J 6 3.85 -0.78 4.63 33-40 days 3.29 (ETp = 3. 1.86 05 cm) -1.30 5.15 1 1.47 0.59 0.88 1.77 0.61 -0.91 2.38 2 3.29 -0.16 3.45 2.92 1.29 -0.92 4.21 3 5.17 0.36 4.81 3.23 0.39 1.11 4.06 4 5.11 1.20 3.91 2.72 0.79 1.60 3.51 5 1.54 -3.56 5.10 2.56 2.03 -3.05 4.59 6 3.29 -1.61 4.90 40-45 days 2.78 (ETp = 1. 0.75 69 cm) -0.24 3.53 1 2.99 1.03 1.96 1,63 0.08 1.28 1.71 2 2.99 0.79 2.20 1.77 0.06 1.16 1.83 3 2.99 0.83 2.16 1.77 0.02 1.12 1.87 4 2.99 1.48 1.51 1.77 0.02 1.20 1.79 5 0 -0.02 0.02 0.77 0.00 -0.77 0.77 6 2.99 0.25 2.74 1.75 0.19 1.05 1.94

PAGE 130

117 Table 19 — continued. Measured water Estimated water Treatmentf Inputt AS§ Depletion ETa Drainage aS§ Depletion^ cm 45-51 days (ETp = 2 .45 cm) 1 6.94 -0.03 6.97 2.47 2.93 1.54 5.40 2 6.94 -0.58 7.52 2.59 4.78 -0.43 7.37 3 6.94 -1.43 8.37 2.59 5.02 -0.67 7.61 4 6.94 -1.44 8.38 2.59 5.25 -0.90 7.84 5 0 -2.42 2.' 42 0.76 0.05 -0.81 0.81 45-71 days (ETp = 9 12 cm) 6 20.00 0.50 19.50 10.00 9.42 0.58 19.42 51-71 days (ETp = 6, .67 cm) 1 11.24 0.52 10.72 6.58 5.85 -1.19 12.43 2 13.06 -0.36 13.42 7.41 6.44 -0.79 13.85 3 13.06 -0.03 13.09 7.20 6.57 -0.71 13.77 4 11.24 0.43 10.81 6.82 5.17 -0.75 11.99 5 0 -0.63 0.63 1.25 0.00 -1.25 1.25 71-85 days (ETp = 3. 72 cm) 1 7.13 0.46 6.67 4.09 2.29 0.75 6.38 2 8.95 -1.50 10.45 4.20 3.84 0.91 8.04 3 7.13 0.39 6.74 4.20 0.81 0.60 6.53 4 7.13 0.20 6.93 4.20 2.13 0.80 6.33 5 7.13 4.06 3.07 2.90 0.22 4.01 3.12 6 0 -3.08 3.08 3.00 2.85 -5.85 5.85 85-98 days (ETp = 3. 78 cm) 1 1.30 -7.64 8.94 3.50 2.63 -4.83 6.13 2 3.12 -3.62 6.74 3.90 2.62 -3.40 6.52 3 1.30 -4.92 6.22 3.48 2.42 -4.38 5.68 4 1.30 -6.35 7.65 3.50 2.59 -4.79 6.09 5 3.12 -1.04 4.16 3.35 0.00 -0.23 3.35 6 0 -1.34 1.34 1.41 0.02 -1.43 1.43

PAGE 131

118 Table 19 — continued. Measured water Estimated water Treatment! Input;}: AS§ Depletion ETa Drainage as§ DepletionH 98-111 days cm (ETp = 3 42 cm) 1 1.78 -0.58 2.36 1.95 0.07 -0.24 2.02 2 3.30 0.61 2.69 2.91 0.07 0.32 2.98 3 3.60 -0.16 3.76 2.56 0.13 0.91 2.69 4 1.78 -0.65 2.43 1.94 0.07 -0.23 2.01 5 3.60 0.91 2.69 2.88 0.00 0.72 2.88 6 0 -0.08 0.08 0.66 0.00 -0.66 0.66 111-115 days (ETp = 1 09 cm) 1 0 -0.72 0.72 0.58 0.00 -0.58 0.58 111-125 days (ETp = 3. 51 cm) 6 3.83 2.69 1.14 1.37 0.04 2.42 1.41 lll-119 days (ETp = 1. 95 cm) 2 3.83 2.22 1.61 1.58 0.19 2.06 1.77 3 3.83 2.31 1.52 1.44 0.14 2.25 1.58 4 3.83 2.08 1.75 1.16 0.16 2.51 1.32 5 3.83 1.64 2.19 1.66 0.23 1.94 1.89 115125 days (ETp = 2. 42 cm) 1 3.83 2.72 1.11 1.96 0.02 1.85 1.98 119125 days (ETp = 1. 56 cm) 2 0 -1.12 1.12 1.56 0.00 -1.56 1.56 3 0 -0.70 0.70 1.53 0.01 -1.54 1.54 4 0 -1.06 1.06 1.38 0.00 -1.38 1.38 5 0 -0.94 0.94 1.56 0.00 -1.56 1.56 fSee p. 28 for description of treatments. ^Rainfall plus irrigation. §Soil profile water content on last day of period minus content on first day. HEstimated ET plus estimated drainage.

PAGE 132

119 H W o •H 4_j QJ rH QJ Q )^ Q) XJ 4-100 a) VI .H e o H jj CO QJ 4J DO o to p, M 4_t 03 QJ }_] (U w 4J (U CO l-j w CO o o 4_j M-l QJ o rH p, QJ QJ Q CS 4j — 1 rt n Mco > t> rrl iH 4J CO 3 C &, O c to •H 0) iH m CO J-1 iH O tfl H •U O H a c QJ 6 J-l CO QJ U H vO VD vD NO NO .H r-l rH rH rH • c CO o u o CO > >-l •H QJ 4_) 4J rH CO 3 O u — rH QJ o •H QJ U QJ a. Q •T3 rH QJ rH 4-1 QJ Cfl e •H u t3 cn c Q) CO c 4-1 CO QJ >-) T3 QJ CO U •H 60 CO 1-1 CO O QJ QJ E O E o rC u 4-) m i+H o 0) 0) e 4J o 4J CO MH CO rH QJ 3 1-J O QJ W rH -H CO rH MH a p. o a crt CO c < o •H T) 4-1 Q) C 4-> CO •H CO rH M e CO Cfl 4-1 QJ cn l-l T3 QJ QJ CO QJ a c l-l T3 o c IIH CO CO QJ CM (U • >-l a 3 cn QJ CO 4-1 QJ QJ cn CA) S W 1— -H-C05 CO CO e

PAGE 133

120 balances show a close agreement between the measured and simulated soil profile water depletions. Table 20 shows that the greatest seasonal depletion difference was 2.14 cm which was observed in the stressed treatment. The good agreement of periodic and seasonal water depletions were a result of the good prediction of water content by the simulation model. Comparisons between measured and simulated profile water depths throughout the growing season for the rainfed and well irrigated treatments are presented in Fig. 29. The model gave slightly lower water depths than the measured values for the 40-85 days period. This could mean that the ETa values were overestimated as a result of too high Kc values. Figure 29 shows lower water depths in the rainfed treatment than the well-irrigated treatment 2 from day 85 to the end of the season. Periodic AS values (Table 19) prior to maximum root depth are not equal to the associated water depth differences in Fig. 29. Comparison between measured and simulated water depth data for treatments 3, 4, 5, and 6 are presented in the Appendix. The ETa estimates of treatments 2, 3, and 4 exceeded the ETp rates during the period 40-85 days. During this period the crop had reached full canopy, and the high ETa estimates were the result of Kc values greater than 1. It is interesting that the model predicted lower water content values than those measured during the induced water stress period in treatments 5 and 6. This is evidence for the occasional failure of the rainshelters as mentioned earlier. Seasonal ETa values represented 77, 89, 86, 82, 62, and 70% of the seasonal ETp for treatments 1, 2, 3, 4, 5, and 6, respectively.

PAGE 134

121 SIMULATED Fig. 29. Measured and simulated root zone water content distribution with time under rainfed (treatment 1) and well irrigated (treatment 2) conditions, 'Georgia Jet' sweet potatoes, 1982.

PAGE 135

122 'Georgia Jet' and 'Yellow Jewel' irrigation-use efficiencies calculated from data of Table 18 for irrigated treatments 2, 3, and 4 are presented in Table 21. The lowest efficiency was given by the well-irrigated treatment 2. In this treatment, the second and fourth irrigations were followed by several continuous rainfall events, and the irrigation water already stored increased the drainage losses. Note that the lower seasonal irrigation in treatment 4 gave the highest irrigation-use efficiency. The average efficiency (51%) of treatments 2, 3, and 4 was close to that previously estimated by the ratio of the two production coefficients (45%)

PAGE 136

123 Table 21. 'Georgia Jet' and 'Yellow Jewel' sweet potato irrigationuse efficiency. 1982. Treatment Irrigation ETa increase Irrigation-use efficiency cm 2 10.92 4.90 3 7.35 3.71 4 3.64 2.09 AVERAGE 7.90 3.57 0.45 0.50 0.57 0.51

PAGE 137

GENERAL DISCUSSION Crop Water Production Function In general, the yields of the five crops In the present studywere linearly related to both irrigation and estimated evapotransplratlon. The production functions reflected the nature of the crop, weather, and crop management practices. The values of the coefficients were In the range found by other researchers. The production functions for oat dry matter were obtained over a relatively short winter season of vegetative growth only. The corn study was the only one where a fertilizer variable (nitrogen) was Included, and there was a marked Interaction with water management. The soybean experiment was the only one of the five experiments where there was good evidence for a differential growth stage response to ETa deficit. Irrigation treatments had no significant effect on pod yields of the peanut genotypes under study because of the wet season. The ralnsheltered water management treatment on four peanut genotypes allowed the identification of ETa production functions. Marked varietal response differences to seasonal evapotranspiration and irrigation amounts was observed in the sweet potato experiment. The high ETa function for marketable yield was given by 'Georgia Jet' while 'Yellow Jewel' showed little response to irrigation or to ETa. The latter was true even on the two ralnsheltered treatments. On the other hand, 'Georgia Jet' yields were not reduced by ralnsheltering and these data points were not on the linear response line associated 124

PAGE 138

125 with the rainfed and three irrigated treatments. This unexpected result was similar to that found in the soybean experiment, where the rainfed yields were unusually high relative to irrigated yields. Suggested factors contributing to the sweet potato results included: uncertainty of root depth and level of water stress experienced by the plants, possible underestimation of ETa, and the underestimation of rainfall input because of the occasional ineffectiveness of the rain shelters. Except for the rainshelter factor, these factors may apply also to the soybean results. In addition, the moderate water deficit incurred during early season soybean growth could have preconditioned the plants to more effectively resist the effect of water stress in later growth stages. Spatial variability of the plow pan condition present in the experimental plots contributed to a high coefficient of variation (22%) for the rainfed plots relative to the well-irrigated plots (10%). Thus, there was a greater possibility of sampling bias toward the more uniform and higher yielding areas on the rainfed treatment. From the standpoint of economic analysis, the irrigation production function is the most useful to the irrigator. The analysis is easier when the function is linear. The seasonal level of irrigation which would be needed on a particular crop in order to pay the annual fixed and operating costs of irrigation can be obtained from the production function converted to dollar return per dollar cost of irrigation. In Florida, the practice of multiple cropping increases the economic justification for irrigation.

PAGE 139

126 Water Management Strategy The water management strategy used for the optimally irrigated treatment experiments was to apply water in small amounts at frequent intervals to meet the ET demand and timed to avoid yield-limiting water stress. In most cases, growth proceeds completely unimpaired and crop yield is maximal only when water potential remains high throughout the growing season (Slatyer, 1969; Hsiao, 1973). Rawlins and Raats (1975) in their paper on the prospects of high-frequency irrigation concluded that converting from low to high-frequency irrigation can save water by decreasing deep percolation. Our basic management strategy on well-drained sandy soils of Florida was the replenishment of a part rather than the full soil water deficit in the root zone in order to leave a portion of the depleted profile available to store intermittent rain. The expected result is a more efficient utilization of irrigation and rainfall while minimizing the transport of water, nutrients, and pesticides to depths greater than plant roots. Variable results were obtained in using our water management strategy depending on the combination of rainfall occurrences and irrigation inputs. Irrigation-use efficiency varied from a low of 34% in the peanut experiment to a high of 89% in the oat experiment. In the corn, soybean, and sweet potato experiments, the respective efficiencies were 75, 58, and 51% for the better subtreatments These data support the view that irrigation of well-drained sandy soils in Florida can be expected to increase drainage loss. It is

PAGE 140

127 a question of scheduling irrigation in a manner which will minimize drainage losses. Computer-Aided Irrigation Scheduling In all of the water management experiments, the current simulation model provided satisfactory estimates of water contents in the root zone. The model could be used to schedule irrigation. Nevertheless, it is important to understand the critical need for a delicate balance of those parameters which characterize the soil-water-plant-system. Assuming that ETp is accurately estimated from meterological data, on one side of the balance are soil-water factors: field capacity, water redistribution time, and root depth. On the other side are the waterplant and soil-water factors affecting ETa: water availability coefficient and crop coefficient. The mix of parameter values which we used gave a satisfactory balance, but the validity of the estimated ETa and drainage values is unknown. A measure of drainage will give ETa through equation 11. On the other hand, a reasonably good test of the validity of ETa can be obtained during periods when there was strong evidence that drainage from the root zone was negligibly small. A selection of periodic water balance data from all experiments was made where the following criteria were met: negligible drainage, full crop canopy, no plant water stress, and best subplot management conditions. The comparative "measured" and estimated daily ETa values as well as calculated ETp span all seasons of 1982 and are shown in Table 21. In the oat lysimeter experiment ETa estimates by the simulation model were not provided. The real daily ETa rates

PAGE 141

128 Table 28. Comparative "measured" and estimated daily ETa as well as daily ETp, under the criteria of negligible drainage, full crop canopy, no plant water stress, and best subplot management conditions. Time Estimated Measured! Estimated§ Datef Crop Treat. period ETa ETa ETp days cm day Jan. 1 Oat 1-5 64-75 0.42 0.23 12 Oat 3 75-95 0.47 0.28 24 Oat 3 95-98 — 0.60 0.33 27 Oat 3 98-101 0.61 0.28 Apr. 29 Corn 2 57-68 0.41 0.41 0.40 May 22 Com 3 84-88 0.43 0.40 0.45 July 4 Com 3 127-130 0.41 0.37 0.53 Aug. 9 Peanut 2 94-100 0.49 0.48 0.45. 16 Peanut 2 101-106 0.53 0.49 0.49 19 Peanut 2 106-108 0.37 0.26 0.34 Oct. 11 Soybeans 6 99-106 0.30 0.39 0.29 18 Soybeans 6 106-113 0.26 0.28 0.27 24 Soybeans 6 113-119 0.22 0.23 0.25 Aug. 20 Sweet potato 2 40-45 0.35 0.44 0.34 Oct. 21 Sweet potato 2 98-111 0.21 0.22 0.26 Nov. 7 Sweet potato 2 119-125 0.26 0.19 0.26 fDate at midpoint of time period. ^"Measured" in the sense that measured inputs and neutron measured soil water contents were used with equation 11 to calculate ETa, given drainage equal to zero. Oat data were an exception, ETa was measured in lysimeters. §Estimated ETp by Penman's method.

PAGE 142

129 measured were much higher than the daily potential rates as estimated by Penman's method. As mentioned earlier, the high ETa values could be attributed to the advection of energy from dry areas surrounding the lysimeters. In the four field-plot water management experiments, there was a close agreement between the estimated and "measured" daily ETa rates. In general, these values were also in close agreement with daily ETp rates. The results indicate that the simulation model is a useful research tool to evaluate the water balance consequences of water mangement practices for Florida soils and climate conditions. Future Research Needs The development of water and other management practices to achieve more crop yield per unit of available water represents the major challenge of agricultural production in many regions of the world. The findings in the present study give added visibility to research needs in water mangement for crop production. 1. Development and testing of practical field methodologies for measuring the drainage component of water balance. With additional measured components, water inputs and soil water content, actual evapotranspiration can be obtained. 2. Use of lysimeters along with field plots to provide better measures of water balance components and field variability. 3. Development and testing of simulation models for scheduling irrigation, predicting fertilizer and pesticide movement in the soil profile, and simulation of crop growth.

PAGE 143

130 A. A more extensive field water mangement experimentation is needed in order to provide ETa production functions and to determine the relative importance of water stress at different plant growth stages. What are the effects on these functions of limiting factors such as fertilizers, tillage, soil, plant population, insects and diseases? 5. What are the relationships between irrigation practices, fertilizer and pesticide use, and soil water pollution problems? 6. There is a need for more knowledge on plant water relations for a wider variety of crop plants. Research on the relationship of leaf water potential and plant growth and yield is plagued with problems caused by little understood processes of plant aging, osmotic regulation, and plant stress history. 7. Development of more extensive soil water data to include spatial variability in soil properties as well as in water applications. Additional data are needed on soil water retention and movement in relation to water redistribution rates, hydraulic conductivity and water flow to roots. 8. Development of quantitative data on root growth and distribution patterns under a variety of soil conditions. 9. Testing of high frequency irrigation strategies based on a partial rather than complete refilling of the depleted profile of sandly soil in humid regions.

PAGE 144

CONCLUSIONS Based on results obtained from one field lysimeter and four field plot water management experiments, the following conclusions can be drawn. 1. Dry matter and marketable yields were linearly related to seasonal evapotranspiration as well as seasonal irrigation amounts. Even though in two experiments (corn and peanut) the linear relationships were defined only through three water management treatments, this finding is in total accordance with most of the results reported in the literature. In the present study there was only one case where the data suggested a possible curvilinear (concave downward) irrigation response function — corn under nitrogen stress. Thus, the highest levels of irrigation were as efficiently used by the crop as the lower levels. The conclusion is that agricultural production must be based on a total utilization of water up to the maximum ET demand of the atmosphere. 2. The linear ETa production function is not a unique relationship as evidenced in the corn and soybean experiments. The relationship is a family of linear functions depending on nutrient stress and timing of water deficits in relation to plant growth stages. 3. Water production functions and the associated irrigation-use efficiencies are considerably reduced when nutrient stresses, such as the nitrogen deficiency in the corn experiment, occur during the growing season. This finding emphasizes the fact that the impact of other growth factors affecting production must be well understood in irrigated agriculture. 131

PAGE 145

132 4. Varietal response differences to water management are reflected in the various water production functions obtained in the peanut and sweet potato experiments. Maximum return from irrigation, allocation and use of water resources in agriculture, requires the total knowledge of the impact of crop varieties on water production functions 5. The water management strategy of applying small amounts at frequent intervals and the partial replenishment of the water depleted soil profile minimizes but does not eliminate drainage and nutrient losses from the root zone. Irrigation of well-drained soils in Florida can be expected to increase deep percolation. It is a question of using irrigation scheduling practices which minimize these losses and, at the same time, meet the ET needs of the crop. More research is needed in this area. 6. The current water balance simulation model (NITROSIM) provided good estimates of the changes in soil-water storage in the root zone. This means that there was a good match of the input values for parameters which characterized the system. Nevertheless, the estimated actual evapotranspiration was verified throughout the year for various short time periods (3-20 days) when it was evident that drainage was negligible. The simulation model represents a potential research tool for scheduling irrigation as well as for the evaluation of the water balance consequences of water mangement practices for Florida soil and climate conditions.

PAGE 146

APPENDIX

PAGE 147

o ^0 ^Hid3a nios 134

PAGE 148

U5 Table 23. Crop coefficients (Kc) used in the corn experiment. Crop coefficients in treatment number 1 2 3 Time period ^0* "it days 013 0.50 0.50 0.50 0.50 0.50 0.50 1326 0.54 0.52 0.60 0.58 0.59 0.58 2639 0.60 0.56 0.70 0.66 0.70 0.64 3952 0.68 0.62 0.85 0.78 0.83 0.76 5265 0.78 0.79 1.00 0.88 0.96 0.84 6578 0.82 0.76 1.10 0.96 1.00 0.92 7891 0.70 0.68 1.15 0.94 0.96 0.92 91-104 0.60 0.60 1.00 0.88 0.88 0.84 104-117 0.54 0.54 0.90 0.80 0.80 0.78 117-132 0.44 0.44 0.80 0.70 0.70 0.68 tNitrogen sufficient conditions. $Ni trogen stress conditions.

PAGE 149

136 Table 24. Crop coefficients (Kc) used in the peanut experiment. Time period Crop coefficients in treatment number 1 2 3 days 014 0.56 0. 56 0 .DO 1428 0.62 0.62 0.62 2842 0.70 0.74 0.74 4256 0.88 0.88 0.88 5670 1.02 1.02 1.02 70-112 1.10 1.10 1.10 112-126 1.00 1.00 1.00 126-139 0.90 0.90 0.90

PAGE 150

137 Table 25. Crop coefficients (Kc) used in the soybean experiment. Crop coefficient in treatment number 1 9 z 4 5 6 days Uli (J 5U 0.50 0. 50 U J u 1 JZD U DH 0.64 0 64 U Oh U OH U OH o ^ on ZDjy 0. 80 0.80 U OU U OU U OU 0.84 0.84 n QA n QA 5265 0.86 0.88 0.88 0.94 1.02 1.04 6578 0.88 0.92 0.92 0.96 0.94 1.10 7891 0.90 0.94 1.02 1.02 0.86 1.10 91-103 0.80 0.86 1.04 1.04 0.84 1.04 103-108 0.75 0.80 1.00 1.00 0.80 1.00 108-113 0.60 0.70 0.90 0.85 0.65 0.95 113-118 0.40 0.60 0.85 0.80 0.45 0.90 118-123 0.70 0.65 0.75 123-127 0.50 0.50 0.55

PAGE 151

138 Table 26. Crop coefficients (Kc) used in the sweet potato experiment. Crop coefficients in treatment number Time period 1 2 3 4 5 6 days u— c J 0.40 0.40 0.40 0.40 0.40 0.40 515 0.45 0.45 0.45 0.45 0.45 0.45 1524 0.60 0.60 0.60 0.60 0.60 0.60 2433 0.80 0.90 0.90 0.90 0.90 0.90 3339 0.90 1.00 1.00 1.00 1.00 1.00 3950 1.00 1.05 1.05 1.05 1.05 1.05 5060 1.05 1.10 1.10 1.10 1.10 1.10 6070 1.10 1.15 1.15 1.15 1.15 1.15 7080 1.10 1.15 1.15 1.15 1.15 1.15 8090 1.10 1.10 1.10 1.10 1.10 1.10 90105 1.05 1.05 1.05 1.05 1.05 1.00 105125 1.05 1.00 1.00 1.10 1.00 1.00

PAGE 152

139 • \0 O CO CJN in rH in o ed > rH o o CM CO 1 — 00 00 TJ o >H CM CM CM CM CM CM CM CM U 0 H 00 in 00 O CJN fO CJN O rH CM in CM CM CM 00 CM o 0) a 1-4 o CM CM CO CO eg CO CM CO CO H •H > tn (U 4H u-1 00 rH a\ CO vO in o CJN c & CS vT 00 CO Cvl 00 in in •rl OJ cd ro m CO CO CO CM CM CM CM o • 00 CO 00 00 in eg CO rH o(30 cs 00 in rH ON CJN 4J me < CM -dCO > 0) CM vO in cr. rrH ro <• c in o 00 CO O CJN vO •TS 3 (U CN ta rH o ON CM o in 00 rH 00 4J u XI in rH in 00 in O 0) CU (d •i ro CM CM CM CM rH CM eg CM > > o rH z in rH rH CM CM rH in O td 1 c in vO CM VO rH rH -) rH rH CM rH rH CM CM rH rH c ON rH o V-i a (U rH o CO CO a rH td o 00 O ON rH CM CM O CO o O in o > ON ^ Cd CO rH CO -3in VO 00 ON H O CMCMCMCMCMCMCMCMCMCMCMCMCM CMCMCMCMCMCMCMCMCMCM CM eg CM rHrHrHrHrHrHrHrHrHrHCMCMCMCMCMCM

PAGE 153

140 vO o > eg o o 2 CN iH CN CN CN rH 00 rH vO 4-1 00 iH CO a o 00 00 vO a iH iH CM CO •oCN CO cn CO CM 00 o iH bo 00 CN O as < •^ in a> vO CM iH o in 3 CO 00 CO 00 c CO vO 00 <-> CN m u-l in 00 tH CN in iH H >> 1 CM a\ in u-1 in .H m 1^ o z CN CM CN eg CN 0) >% c CO 00 CJS o H O CN CM CM CM CO CM

PAGE 154

lAl Table 28. Measured and simulated profile water depths with time for three water management treatments on corn, under nitrogen stress conditions. Water depth in treatment number Root 1 2 3 UCL L. d lyd y Lieu ULl D J.LI1* Meas oxm. Meas sxm. Mar 16 20 20 1.00 0.60 cm 1.00 0. 60 1.00 0. 60 Apr. 9 43 48 6.90 5. 39 6.83 5. 36 6.83 5.36 23 57 64 3.75 2.88 5.24 4. 20 5.20 4. 32 May 4 68 76 6.73 6.02 4.87 4.36 7 71 84 3.55 3.59 8.05 6. 71 3. 73 3.44 9 73 88 3.68 3. 78 10 74 88 6.16 6.24 iz 1 D yz J. Jo J. 39 6 79 6.52 17 81 104 2.49 2.56 9.49 8.98 7.90 6.45 20 84 112 7.72 10.70 8.47 8.47 24 88 116 6.35 8.52 12.32 13.05 12.37 12.38 27 91 116 10.74 11.38 10.53 10.26 June 3 98 120 7.94 11.36 9.94 11.12 16 111 120 9.84 10.12 12.00 10.41 10.25 10.43 29 124 120 11.07 10.57 13.97 10.74 10.69 10.74 July 2 127 120 10.82 9.27 8.63 9.26 5 130 120 9.72 8.49 8.91 8.54 8.13 8.54

PAGE 155

142 W 4J r* \-t di 1 u (d 0) u [ } (U a tit c i 0) td w ll I B *rH [ J CiJ J— f r\ w *rH *H w CO <^ (11 w (11 *H rri ^ 1 > > O d) 1 1 1 w O *. M 0-1 H OJ ID -J i~H CO es n "3 "H (U CO >^ o C to CT3 ^ M o CJ cn td 0) c o CM cd H o cx O 0) Pi T3 Q td Q O CN O 00 00 O 00 1^ 3in to vO 00 NO iH CN 00 NO in I— 1 1 — 1 O 00 o r~ NO CM O ON o o rH in O CN 00 00 in 00 O 00 00 00 o rH rH CM CN CO <• o

PAGE 156

143 n 4-1 CO 0) u c 0) CO 4 (U cd O <4-l M O m J3 4-1 0) •a 0) 4-1 CO <4-l Cd O 4J u o &. o. (U d) 4-1 (U CO 5 iH W 3 e •H 4J 05 OJ C CO •H oo (U O U O in rH rH CJN CO O 00 O o O o o o CN -ao rH CO VD 00 00 00 00 00 00 rH rH rH rH rH rH rH rH rH rH oco o in rH rH in 00 rH in ON in CN CO > iH bO a. 4J > 3 3 QJ O i-i <: CO o

PAGE 157

144 Table 31. Soil water parameters, Kendrick fine sand corn experiment. cm cm cm 014 0.14 0.119 0.077 0.030 1430 0.11 0.080 0.057 0.025 3070 0.10 0.092 0.063 0.022 70-102 0.13 0.080 0.059 0.030 102-120 0.18 0.134 0.095 0.040 tinfil tration water content, twater content at field capacity. §Residual water content obtained from 6 =[(9pp 6 )/1.7]+ 0 llWater content at 15 bar. "'^

PAGE 158

145 Table 32. Soil water parameters, Kendrick fine sand, peanut experiment. Depth ^15^ cm 3 -3 cm cm 014 0.14 0.11 0.077 0.030 1430 0.11 0.09 0.063 0.025 30-114 0.10 0.07 0.050 0.022 114-134 0.13 0.08 0.056 0.022 134-150 0.20 0.14 0.095 0.030 tinf iltration water content. tWater content at field capacity. §Residual water content obtained from 9^^ ^^6^^ 6-|^^)/1.7]+ B IfWater content at 15 bar.

PAGE 159

146 Table 33. Soil water parameters. Lake fine sand, soybean experiment. Depth 9pc^ ^15' cm 3 -3 cm cm 030 0.12 0.094 0.056 0.030 3058 0.11 0.075 0.047 0.028 5890 0.11 0.077 0.043 0.020 90-118 0.11 0.076 0.042 0.020 118-142 0.12 0.072 0.044 0.025 142-150 0.16 0.104 0.060 0.030 tinf iltration water content, twater content at field capacity. §Residual water content obtained from 9 =1(9 9 )/2.46]+ 9 "Water content at 15 bars. -"-^

PAGE 160

147 Table 34. Soil water parameters. Lake fine sand, sweet potato experiment Depth Ost 61511 cm 3 -3 cm cm 014 0.145 0.094 0.056 0.030 1438 0.130 0.084 0.052 0.030 3898 0.120 0.075 0.045 0.025 98-134 0.125 0.079 0.044 0.020 134-150 0.130 0.077 0.046 0.025 150-180 0.135 0.082 0.048 0.025 tinf iltration water content. iWater content at field capacity. §Residual water content obtained from 9 =1(6,^^ 9^t.)/2.46]+ 6 IfWater content at 15 bars.

PAGE 161

LITERATURE CITED Ambak, K. 1982. Effect of water management on yield response of early and mid season soybeans. M. S. Thesis, University of Florida, Gainesville, Florida. Arkley, R. J. 1963. Relationship between plant growth and transpiration. Hilgardia. 34:559-584. Barrett, J. W. H., and G. V. Skogerboe. 1980. Crop production functions and the allocation and use of irrigation water. Agric. Water Manage. 3:55-64. Bierhuizen, J. F., and R. 0. Slatyer. 1965. Effect of atmospheric concentration of water vapour and CO2 in determining transpiration-photosynthesis relationships of cotton leaves. Agric. Meteorol. 2:259-270. Black, C. A. 1966. Crop yields in relation to water supply and soil fertility. In W. H. Pierre, D. Kirkhan, J. Pesek, and R. Shaw (eds.) Plant environment and efficient water use. Am. Soc. Agron. Madison, Wisconsin. Burns, I. G. 1980. A simple model for predicting the effects of leaching of fertilizer nitrate during the growing season on the nitrogen fertilizer need of crops. J. Soil Sci. 31:175185. Denmead, 0. T. and R. H. Shaw. 1960. The effect of soil moisture stress at different stages of growth on the development and yield of corn. Agron. J. 52:272-274. DeWit, C. T. 1958. Transpiration and crop yields. Institute of Biol, and Chem. Res. on Field Crops and Herbage, Wageningen, The Netherlands. VErsl. Landbouwk. Onderz. No. 64.6. Doorenbos, J., and A. H. Kassam. 1979. Yield response to water. FAG Irrigation and Drainage Paper 33, FAO, Rome. Doss, D. B., R. W. Pearson, and H. T. Rogers. 1974. Effect of soil water stress at various growth stages on soybean yield. Agron. J. 66:297-299. Fischer, R. A. 1979. Growth and water limitation to dryland wheat yield in Australia: A physiological framework. J. Aust. Inst. Agric. Sci. 45:83-94. 148

PAGE 162

149 Garrity, D. P., D. G. Watts, C. Y. Sullivan, and J, R. Gilley. 1982a. Moisture deficit and grain sorghum performance: Effect of genotype and limited irrigation strategy. Agron. J. 74:808-814. and 1982b. Moisture deficits and grain sorghum performance: Evapotranspiration-yield-relationships. Agron. J. 74:815-820. Geiser, K. M. D. C. Slack, E. R. Allred, and K. W. Stange. 1982. Irrigation scheduling using crop canopy-air temperature difference. Transaction ASAE 25:689-694. Hammond, L. C., R. B. Campbell, and E. D. Threadgill. 1981a. Irrigation management strategies for humid regions. In Irrigation Scheduling for Water and Energy Conservation in the 80' s. Proc. Irrigation Scheduling Conf., ASAE, Chicago. Dec. 14-15. R. S. Mansell, W. K. Robertson, J. T. Johnson, and H. M. Selim. 1981b. Irrigation efficiency and controlled root zone wetting in deep sands. Water Research Center, University of Florida Pub. No. 52, Gainesville, Florida. Hanks, R. J., H. R. Gardner, and R. L. Florian. 1969. Plant growthevapotranspiration relations for several crops in the Central Great Plains. Agron. J. 61:30-34. Hiler, E. A., T. A. Howell, R. B. Lewis, and R. P. Boos. 1974. Irrigation timing by the stress day index methods. Transaction ASAE 17:393-398. Hillel, D. and Y. Guron. 1973. Relation between evapotranspiration rate and maize yield. Water Resources Res. 9:743-748. Hsiao, T. C. 1973. Plant responses to water stress. Ann. Rev. Plant Physiol. 24:519-570. Jackson, R. D., R. J. Reginato, and S. B. Idso. 1977. Wheat canopy temperatures: A practical tool for evaluating water requirements. Water Resources Res. 13:651-656. Jensen, M. E. 1968. Water consumption by agricultural plants. PP. 1-22. In T. T. Koslowski (ed.) Water deficits and plant growth, 2. Academic Press, New York. and J. T. Musick. 1960. The effect of irrigation treatments on evapotrnaspiration and production of sorghum and wheat in the Southern Great Plains. Int. Soil Sci. Soc. 7th Cong. Trans. 1:386-393. ^' C. N. Robb, and C. E. Franzoy. 1970. Scheduling irrigation using climate-crop-soil data. Am. Soc. Civ. Eng. Proc, J. Irrig. Drain. Div, 96:25-38.

PAGE 163

150 Jones, S. T. 1961. Effect of irrigation at different levels of soil moisture on yield and evapotranspiration rate of sweet potatoes. Amer. Soc. Hort. Sci. Proc. 77:458-462. Monteith, J. L. 1965. Radiation and crops. Exp. Agric. 1:241-251. Musick, J. T., D. W, Grimes, and G. M. Herron. 1963. Water management, consumptive use and nitrogen fertilization of irrigated winter wheat in western Kansas. USDA Prod. Res. Rpt. No. 75, Washington, D. C. L. L. New, and D. A. Duseck. 1976. Soil water depletionyield relationships of irrigated sorghum, wheat, and soybeans. Trans. Am. Soc. Agr. Eng. 19:489-493. Palacios, V. E. 1977. Introducion a la teoria de la operacion de distritos y systemas de riego. Rama de Riego y Drenaje Colegio de Postgraduados Chapingo, Mexico. Pallas, J. E., Jr., J. R. Stansell, and T. J. Koske. 1979. Effects of drought on Florunner peanuts. Agron. J. 71:853-858. Penman, H. L. 1948. Natural evaporation from open water, bare soil and grass. Proc. Roy. Soc. London, Ser. A. 193:120-146. Persad, S. 1982. Root growth and distribution of six corn cultivars in an Arenic Paluedult as influenced by irrigation. M. S. Thesis University of Florida, Gainesville, Florida. Phene, C. J., and 0. W. Beale. 1976. High-frequency irrigation for water management in humid regions. Soil Sci. Soc. Am. J. 40: 430-436. Rao, P. S. C., J. M. Davidson, and L. C. Hammond. 1976. Estimation of nonreactive and reactive solute front locations in soils. In Residual management by land disposal. Proc. Hazardous Waste Residue Sump. Tucson, Arizona. and R. E. Jessup. 1981. Simulation of nitrogen behavior in the root zone of cropped land areas receiving organic wastes. In M, J. Frissel and J. A. van Veen (eds.) Simulation of nitrogen behavior in soil-plant systems. Pudoc, Wageningen, The Netherlands. Rawlins, S. L. and P. A. C. Raats. 1975. Prospects for high frequency irrigation. Science. 188:604-609. Rhoads, F. M. 1981. Plow layer soil water management and program fertilization on Florida Ultisols. Soil Crop Sci. Soc. Fla. Proc 40:12-14.

PAGE 164

151 Ritchie, J. T. 1972. Model for predicting evaporation from a row crop with incomplete cover. Water Resources Res. 8:1204-1213. Sajjapongse, A., and Y. C. Roan. 1982. Physical factors affecting root yield of sweet potato [ Ipomoea batatas (L.) Lam]. In R. L. Villareal and T. D. Griggs (eds.) Sweet Potato. Proc. of First Int. Symp. Asian Vegetable Res. and Dev. Center Shanhua, Rainan, Taiwan, China. Shipley, J., and C. Regier. 1975. Water response in the production of irrigated grain sorghum. High Plains of Texas. Tex. Agric. Exp. Stn. MP-12-2. and 1976. Corn yield response to limited irrigations, High Plains of Texas. Tex. Agric. Exp. Stn. Prog. Rpt. PR-3379C. Sinclair, T. R., C. B. Tanner, and J. M. Bennett. 1984. Crop wateruse efficiency and increased yields. Bioscience 34:36-40. Sionit, N. and P. J. Kramer. 1977. Effect of water stress during different stages of growth of soybean. Agron. J. 69:274-278. Skogerboe, G. V., J. W. H. Barret, B. J. Treat, and D. B. McWhorter. 1979. Potential effects of irrigation practices on crop yields in Grand Valley. U. S. Environmental Protection Agency, Ada, Olkahoma. EPA-600/2-79-149 Slatyer, R. 0. 1969, Physiological significance of internal water relations. In J. D. Eastin, F. A. Haskins C. Y. Sullivan, and C. H. M. van Bavel (eds.) Physiological aspects of crop yields. Am. Soc. Agron., Madison, Wisconsin. Smajstrla, A. G. 1982. Irrigation management for the conservation of limited water resources. Project completion report. Agric. Eng. Dept., IFAS, University of Florida, Gainesville, Florida. ai^d G. A. Clark. 1981. Water stress effects on water use and yield of soybeans. Soil Crop Sci. Soc. Fla. Proc. 41:178181. Stegman, E. C. J. T. Musick, and J. I. Stewart. 1980. Irrigation water management. In M. E. Jensen (ed.) Design and operation of farm irrigation systems. Am. Soc. Agric. Ent., St. Joseph, Michigan. 18:764-809. Stewart, J. I., and R. H, Hagan. 1973. Functions to predict effects of crop water deficits. Proc. ASCE, J. Irrig. Drain. Div. 99 (IR4) :421-439.

PAGE 165

152 Stewart, J. I., R, H. Hagan, and W. 0. Pruitt. 1976. Production functions and predicted irrigation programs for principal crops as required for water resources planning and increased water use efficiency. Final Rpt. U. S. Dept. of the Interior, Bur. of Reclamation, Washington, D. C. 5 R. J. Hanks, J. P. Riley, R. E. Danielson, W. T. Franklin, and E. B. Jackson. 1977. Optimizing crop production through control of water and salinity levels in the soil. Utah Water Res. Lab., Utah State Univ., Logan, Utah. Pub. No. PRWG151-1. Tanner, C. B., and T. R. Sinclair. 1983. Efficient water use in crop production: • Research and re-search. In H. Taylor, W. Jordan, and T. R. Sinclair (eds.) Limitations to efficient water use in crop production 1:1-46. Am. Soc. Agron. Madison, Wisconsin. Tennant, D. 1975. A test of a modified line intersect method of estimating root length. J. Ecol. 63:995-1001. Varnell, R. J., H. Mwandamere, W. K. Robertson, and K. J. Boote. 1976. Peanut yields affected by soil water, no-till, and gypsum. Proc. Soil Crop Sci. Soc. Fla. 35:56-59. Verasan, V., and R. E. Phillips. 1978. Effect of soil water stress on growth and nutrient accumulation in corn. Agron. J. 70: 613-618. Viets, F. G., Jr. 1962. Fertilizers and the efficient use of water. In A. G. Norman (ed.) Advances in agronomy. Vol. 14:233-264. Academic Press, New York. • 1966. Increasing water use efficiency by soil management. In W. H. Pierre, D. Kirkham, J. Pesek, and R. Shaw (eds.) Plant environment and efficient water use. Am. Soc. Agron., Madison, Wisconsin. Vittum, M. T., and W. J. Flocker. 1967. Vegetable crops. In R, M. Hagan, H. R. Raise, and T. W. Edminster (eds.) Irrigation of agricultural lands. Am. Soc. Agron., Madison, Wisconsin. 34: 674-685. Yaron, D. 1971. Estimation and use of water production functions in crops. J. Irrig. Drain. Div., Am. Soc. Civ. Eng. 97:291303.

PAGE 166

BIOGRAPHICAL SKETCH David Riestra-Diaz was born in Acapulco, State of Guerrerro, Mexico, on February 17, 1948, He received the degree in Agronomic Engineering at the Escuela Superior de Agricultura, Universidad Autonoma de Guerrerro in 1973. During the period of 1973 to 1974, he was attached to the Water Management District Number 11, Alto Rio Lerma, Mexico, as an Extension Officer. David Riestra-Diaz began his graduate studies at Colegio de Postgraduados Chapingo, Mexico, in the field of drainage and irrigation in February, 1975. In January, 1977, he was awarded the degree of Master of Science. He was hired by the Colegio de Postgraduados Chapingo, Mexico, to carry on research on water management of field crops at the Research and Education Center of Veracruz, Mexico. He entered the University of Florida in Fall of 1980, on a CONACYT Scholarship to pursue graduate studies leading to a Ph. D. degree in soil science. He married Maria Guadalupe Monrroy Arce in 1972, and they have three daughters, Nancy Guadalupe, Monica, and Norma Fabiola. 153

PAGE 167

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Luther C. Hammond, Chairman Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. S. C. Rao Associate Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. An,. <^.5^^fJcL Allen G. Smaj stria / Associate Professor of Agricultural Engineering

PAGE 168

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. K. J. Boote Associate Professor of Agronomy I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. c^^/^ ^ /^^^^ Z.^. Beiinett Assistant Professor of Agronomy This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. April 1984 Dean for Graduate Studies and Research