Growth of young 'Hamlin' orange trees as influenced by microsprinkler irrigation, fertilization, and nursery tree type

MISSING IMAGE

Material Information

Title:
Growth of young 'Hamlin' orange trees as influenced by microsprinkler irrigation, fertilization, and nursery tree type
Physical Description:
xi, 173 leaves : ill. ; 28 cm.
Language:
English
Creator:
Marler, Thomas E., 1959-
Publication Date:

Subjects

Subjects / Keywords:
Oranges -- Growth   ( lcsh )
Oranges -- Roots   ( lcsh )
Horticultural Science thesis Ph.D
Dissertations, Academic -- Horticultural Science -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1988.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
Thomas E. Marler.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001469572
oclc - 20857955
notis - AGY1265
sobekcm - AA00004816_00001
System ID:
AA00004816:00001

Full Text












GROWTH OF YOUNG 'HAMLIN' ORANGE TREES AS INFLUENCED
BY MICROSPRINKLER IRRIGATION, FERTILIZATION, AND
NURSERY TREE TYPE








By

THOMAS E. MARLER


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

1988
















ACKNOWLEDGEMENTS


Sincere appreciation is extended to Dr. F. S. Davies, chairman

of the supervisory committee, for his supervision of this project

and guidance in preparation of this manuscript. Appreciation is

also extended to Dr. J. J. Ferguson, Dr. L. K. Jackson, Dr. R. C. J.

Koo, and Dr. A. G. Smajstrla for serving on the supervisory

committee and offering helpful suggestions in the planning and

conducting of this research.

Special thanks are extended to Dr. P. C. Andersen for provision

of equipment and to the UF faculty members whose timely counsel made

the continuation of this project possible.


- ii -


















TABLE OF CONTENTS


Page


ACKNOWLEDGEMENTS . . ii

LIST OF TABLES . . .. v

LIST OF FIGURES . . viii

ABSTRACT . . .. x

CHAPTERS

I INTRODUCTION . . 1

II REVIEW OF THE LITERATURE .............. 4


Vegetative Growth and Development
Water Relations .
Irrigation of Mature Citrus ..
Fertilization . .
Young Citrus Tree Care .


of Citrus
* .
* *

* S 0 S




III MICROSPRINKLER IRRIGATION AND GROWTH OF YOUNG
'HAMLIN' ORANGE TREES. I. CANOPY GROWTH AND
DEVELOPMENT ........ *

Introduction .. .......... *
Materials and Methods . .
Results and Discussion . .

IV MICROSPRINKLER IRRIGATION AND GROWTH OF YOUNG
'HAMLIN' ORANGE TREES. II. ROOT GROWTH AND
DISTRIBUTION . .


Introduction . *
Materials and Methods .. ..
Results and Discussion *


V SOIL MOISTURE STRESS AND FOLIAR GAS EXCHANGE
OF YOUNG, FIELD-GROWN 'HAMLIN' ORANGE TREES. .


Introduction . .
Materials and Methods .
Results and Discussion . .


- iii -


LI -









VI GROWTH OF YOUNG 'HAMLIN' ORANGE TREES USING
STANDARD AND CONTROLLED-RELEASE FERTILIZERS .. 114

Introduction . .. .. 114
Materials and Methods .. ... 116
Results and Discussion ........ 119

VII GROWTH OF BARE-ROOTED AND CONTAINER-GROWN
'HAMLIN' ORANGE TREES IN THE FIELD . 125

Introduction . . .. 125
Materials and Methods . .. 126
Results and Discussion ..... .. 129

VIII CONCLUSIONS . . ... .. 141

APPENDICES

A MEAN MONTHLY PAN EVAPORATION AND MAXIMUM AND
MINIMUM AIR TEMPERATURES FOR 1985, 1986, AND 1987 .145

B WHOLE PLANT FRESH AND DRY WEIGHTS AND SHOOT:
RATIO OF YOUNG 'HAMLIN' ORANGE TREES AS RELATED
TO IRRIGATION BASED ON SOIL WATER DEPLETION 146

C REGRESSION EQUATIONS AND COEFFICIENTS OF
DETERMINATION ACCOMPANYING FIGS. 3-7, 3-8,
AND 3-9 . .. . 147

D MEAN AND MAXIMUM LEAF TO AMBIENT AIR TEMPERATURE
DIFFERENCE OF 'HAMLIN' ORANGE TREES IN JUNE 1987
AS INFLUENCED BY SOIL WATER DEPLETION . 148

LITERATURE CITED . . .. 149

BIOGRAPHICAL SKETCH . . .. 173


- iv -
















LIST OF TABLES


Tables Page

3-1. Length of irrigation and amount of water applied at
each irrigation related to soil water depletion for
young 'Hamlin' orange trees .. . 43

3-2. Number of irrigations and cumulative water applied
for young 'Hamlin' orange trees under scheduling
treatments based on soil water depletion .. 46

3-3. Shoot and leaf sizes and expansion rates for three
growth flushes of young 'Hamlin' orange trees as
influenced by irrigation based on soil water
depletion, 1985 . . .. 49

3-4. Shoot and leaf sizes and expansion rates for three
growth flushes of young 'Hamlin' orange trees as
influenced by irrigation based on soil water
depletion, 1986 . . .. 51

3-5. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1985 .... .... .. 53

3-6. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1986 . .. 54

3-7. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1987 . .. 56

3-8. Canopy volume, dry weight, shoot length, leaf area
and trunk cross-sectional area of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion ................. .. 58

4-1. Total root dry weight, dry weight of new roots, and
root volume of young 'Hamlin' orange trees as related
to irrigation based on soil water depletion 74


1









4-2. Lateral dry weight and percentage distribution of
fibrous and total root systems of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1985 .. . 77

4-3. Lateral dry weight and percentage distribution of
fibrous and total root systems of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1986 . . 78

4-4. Lateral dry weight and percentage distribution of
fibrous and total root systems of young 'Hamlin'
orange trees as related to irrigation based on soil
water depletion, 1987 .. 79

4-5. Partitioning of variance into main and interaction
effects for 1987 circular trench profile root growth
variables . . .* 82

4-6. Total root concentration at four depths and two
distances of young 'Hamlin' orange trees as
influenced by irrigation based on soil water
depletion, 1987 .... . 83

4-7. Concentration of fibrous roots at four depths and
two distances from trunk of young 'Hamlin' orange
trees as influenced by irrigation based on soil
water depletion, 1987 . .. 85

4-8. Concentration of non-fibrous roots at four depths
and two distances from trunk of young 'Hamlin'
orange trees as influenced by irrigation based on
soil water depletion, 1987 . 86

4-9. Percentage distribution of root concentration with
depth of young 'Hamlin' orange trees as influenced by
irrigation based on soil water depletion, 1987 87

4-10. Root length density at four depths and two distances
from trunk of young 'Hamlin' orange trees as related
to irrigation based on soil water depletion, 1987 89

5-1. Mean and maximum CO assimilation, transpiration, and
stomatal conductance of young 'Hamlin' orange trees on
18 Oct. 1986 as influenced by soil water depletion 99

5-2. Mean and maximum CO2 assimilation, transpiration,
stomatal conductance, and mean internal CO2
concentration of young 'Hamlin' orange trees in June
1987 as influenced by soil water depletion .... 105

6-1. Effects of controlled-release fertilizers on growth
of young 'Hamlin' orange trees in the field 120


- vi -









6-2. Effects of standard fertilizer rate on growth of
young 'Hamlin' orange trees in the field 122

6-3. Influence of standard fertilizer rate on leaf analysis
of young 'Hamlin' orange trees in the field 123

7-1. Effect of nursery tree type on growth of 'Hamlin'
orange trees after 8 months in the field ... 131

7-2. Effect of removing medium prior to planting on growth
of containerized 'Hamlin' orange trees, 1986 135

7-3. Growth of bare-rooted and container-grown 'Hamlin'
orange trees as influenced by planting procedure,
1987 . * 137


- vii -















LIST OF FIGURES
Figure Page

3-1. The relationship between soil water content and water
potential of Kanapaha sand at the Horticultural Unit 32

3-2. The relationship between neutron probe ratio and
volumetric water content of Kanapaha sand at the
Horticultural Unit ... ... .. 35

3-3. Relationship of leaf area and fresh weight from eight
representative 'Hamlin' orange trees after 8 months
in the field .* * 39

3-4. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, May-Dec. 1985 . 45

3-5. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, May-Dec. 1986 47

3-6. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, April-Dec. 1987 48

3-7. Cumulative percentage of trees in three irrigation
treatments growing over the 1985 season . 60

3-8. Cumulative percentage of trees in three irrigation
treatments growing over the 1986 season . 62

3-9. Cumulative percentage of trees in three irrigation
treatments growing over the 1987 season . 63

3-10. Trunk cross sectional area and canopy volume of young
'Ramlin' orange trees as influenced by microsprinkler
irrigation spray pattern, 1986 . 66

3-11. Trunk cross sectional area and canopy volume of young
'Hamlin' orange trees as influenced by microsprinkler
irrigation spray pattern, 1987 . 67

5-1. Diurnal cycle of photosynthetic photon flux, air
temperature, relative humidity, and vapor pressure
deficit on 18 Oct. 1986 ..... ........... 96

5-2. Diurnal cycle of CO assimilation, transpiration, and
stomatal conductance of young 'Hamlin' orange trees on
18 Oct. 1986 as influenced by soil water depletion 97


- viii -










5-3. Diurnal cycle of photosynthetic photon flux, air
temperature, relative humidity, and vapor pressure
deficit on 12 June 1987 .. . 100

5-4. Diurnal cycle of CO2 assimilation, transpiration,
stomatal conductance, and water use efficiency of
young 'Hamlin' orange trees on 12 June 1987 as
influenced by soil water depletion . .. 101

5-5. Diurnal cycle of photosynthetic photon flux, air
temperature, relative humidity, and vapor pressure
deficit on 13 June 1987 .. .. . 102

5-6. Diurnal cycle of CO2 assimilation, transpiration,
stomatal conductance, and water use efficiency of
young 'Hamlin' orange trees on 13 June 1987 as
influenced by soil water depletion .. 104

5-7. Diurnal cycle of photosynthetic photon flux, air
temperature, relative humidity, and vapor pressure
deficit on 15 June 1987 .. .... . 106

5-8. Diurnal cycle of xylem potential, CO2 assimilation,
transpiration, stomatal conductance, and water use
efficiency of young 'Hamlin' orange trees on 15
June 1987 as influenced by soil water depletion 107

7-1. Effect of 'Hamlin' orange nursery tree type on
increase in trunk cross sectional area and canopy
volume from May 1985 to Dec. 1986 .... 130

7-2. Effect of 'Hamlin' orange nursery tree type on
increase in trunk cross sectional area and canopy
volume from May 1986 to Dec. 1987 ....... .. 133


- ix -


__
















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

GROWTH OF YOUNG 'HAMLIN' ORANGE TREES AS INFLUENCED
BY MICROSPRINKLER IRRIGATION, FERTILIZATION, AND
NURSERY TREE TYPE

By

Thomas E. Marler

December, 1988

Chairman: Frederick S. Davies
Major Department: Horticultural Science (Fruit Crops)

Young tree care is a costly and important part of any citrus

production program. Irrigation, fertilization, and nursery stock

characteristics have a major influence on the success of a young

tree care program. Therefore field experiments were conducted to

study the effects of various microsprinkler irrigation and

fertilization rates and nursery tree types on growth and development

of young 'Hamlin' orange [Citrus sinensis (L.) Osb.] trees.

Three experiments were conducted to determine the effects of

scheduling microsprinkler irrigations at 20 (high frequency), 45

(moderate frequency), and 65% (low frequency) available soil water

depletion on one season's canopy and root growth and leaf gas

exchange. Canopy and root growth were similar for high and moderate

treatments in 2 out of 3 years, but were reduced by the low

treatment, even though the moderate received about 50% less water

than the high treatment. Reduced shoot number and delayed


- x -









initiation of late growth flushes sometimes occurred with moderate

and low treatments, and may have been related to decreases in CO2

assimilation in spring and early summer. Root concentrations were -

greatest between 10 and 30 cm depths, and >90% of the root dry

weight was within 80 cm of the trunk.

Two studies with various fertilizer rates and two studies with

controlled-release fertilizer sources were conducted to determine

the influence of fertilizer rate and source on growth. Growth was

comparable over one season using 0.07-0.09, 0.14-0.18 (recommended),

and 0.22-0.27 kg N/tree/season, implying that adequate rates may

sometimes be lower than recommended. In addition, application of

standard fertilizers (4-5X/season) compared with controlled-release

fertilizers (2X) at the same seasonal rates resulted in similar

growth over 2 seasons, suggesting that controlled-release sources

may sometimes be used as alternatives to standard sources.

Commercial bare-rooted, field-grown nursery trees and

containerized, greenhouse-grown trees were used in three experiments

to compare establishment and initial growth in the field. Bare-

rooted trees remained larger than containerized trees through 2

seasons and initial growth of container-grown trees was highly

variable. Removal of medium prior to planting container-grown trees

improved tree growth in one of two experiments.


- xi -


' '* '















CHAPTER I


INTRODUCTION



Citrus plantings in Florida decreased by more than 75,000

hectares between 1984 and 1986 (Division Plant Industry, 1986), more

than any other 2-year period since the beginning of the Citrus Tree

Census Survey in 1966. This decrease was largely attributable to

the series of freezes since 1982; however, blight, tristeza, and

urbanization have also contributed to the reduction in land area

planted to citrus. Growers estimated in 1983-1984 that 6.1% of

their young trees died each year (Jackson et al., 1986). Estimates

of the number of trees being planted annually in the mid-1980s range

from greater than 3 million (Division Plant Industry, 1986) to 6-10

million (Jackson et al., 1986).

Muraro (personal communication, Citrus Research and Education

Center, Lake Alfred) estimates that more than $20 is required to

maintain a tree in a solid-set planting for the first 4 years in the

flatwood areas, while $24 to $40 per tree is required for resets,

depending on number of resets per hectare. Consequently, young tree

care costs are immense on an industry-wide basis. More efficient

management practices for young citrus trees are important in the

success of any profitable citrus production venture due to the

increased need for cost containment in the Florida citrus industry.


_.__









Vital to the improvement of field management practices is the

understanding of how these practices influence the growth and

physiology of young citrus trees.

Irrigation practices for mature Florida citrus trees are based

on many years of field research, however, information on water

requirements of young citrus trees, irrigation scheduling, and tree

response has not been reported with micro irrigation under field

conditions in Florida. Objectives of the first part of this

research were to investigate the effects of microsprinkler

irrigation scheduling based on soil water depletion on canopy and

root growth of young citrus trees. In addition, the influence of

soil water deficit on gas exchange processes of young citrus trees

was studied. Separate experiments were designed to investigate the

effect of microsprinkler irrigation spray pattern on growth of trees

during the second season in the field.

Fertilization practices for mature Florida citrus are well-

defined and based on years of production records, but fewer studies

have been made on fertilization of young citrus trees. Previous

studies suggest that under many conditions fertilizer rates for

young trees are excessive. Use of controlled-release fertilizers

may reduce nitrogen losses and fertilizer costs by decreasing the

number of yearly fertilizer applications, resulting in reduced labor

and equipment costs. Objectives of the second portion of this

research were to compare growth of young citrus trees using standard

fertilizer applied at recommended seasonal rates and controlled-

release fertilizer applied fewer times per season than the standard

source. A second set of experiments was designed to determine if







3

recommended seasonal fertilization rates could be reduced without a

reduction in young citrus tree growth.

Florida nurserymen have been producing bare-rooted citrus trees

in field nurseries for many years, however, recently many citrus

trees have been produced in containers in greenhouses. While

opinions vary concerning post-plant growth and survival rates of

container-grown and field-grown trees, there have been no

replicated, controlled comparisons of these two nursery tree types

under Florida conditions. The purpose of the third part of this

research was to compare establishment and post-plant canopy and root

growth of 1)containerized, greenhouse-grown and 2)bare-rooted,

field-grown citrus trees. Further experiments were designed to

determine the effect of the removal of potting medium prior to

planting on root and canopy growth of container-grown citrus trees.

Results from these experiments may be useful in improving

management programs for young citrus trees by reducing water and

fertilizer costs and may lead to further research on optimizing

growth and development of young citrus trees in Florida.
















CHAPTER II


REVIEW OF THE LITERATURE



Citrus tree growth and productivity is influenced by many

environmental conditions and cultural factors. Adequate soil

moisture and nutrient levels are two important factors necessary for

optimum tree growth. Moreover, initial growth and development of

newly-planted trees may be influenced by the nursery conditions

under which they were produced. Understanding the influence of

nursery and field management decisions on growth and physiological

processes of young citrus trees is important to maximize early

productivity.

Vegetative Growth and Development of Citrus

Citrus shoot growth is cyclic with two to five flushes per

season, depending on climate (Cooper et al., 1969; Mendel, 1969).

Environmental conditions may alter both number and timing of growth

flushes. Following cessation of shoot elongation the terminal bud

normally aborts and abscises (Schroeder, 1951). Duration of flush

growth is largely dependent upon climate, as cooler spring

temperatures result in longer periods of growth (Cooper et al.,

1963). Both the spring (Cooper et al., 1963) and summer

(Krishnamurthi et al., 1960) flushes have been reported to occur

over the longest time interval.


I







5

Shoot growth is mainly dependent on stored substrates in citrus

(Sinclair, 1984). Available reserves influence shoot length and

number of leaves per shoot (Van Noort, 1969). The spring flush

normally has more shoots than other flushes, but average shoot

length is shorter (Mendel, 1969; Krishnamurthi et al., 1960). Later

flushes have fewer buds breaking and longer average shoot length,

thicker shoots, and larger leaves (Reuther, 1973; Mendel, 1969). A

large proportion of spring shoots do not continue growing during

subsequent flushes (Sauer, 1951). Summer (Mendel, 1969) and spring

(Crider, 1927; Krishnamurthi et al., 1960; Syvertsen et al., 1981)

flushes have each been attributed with having more total growth than

the other growth flushes.

In the spring root growth usually occurs after shoot growth

(Crider, 1927; Cooper et al., 1963; Bevington, 1983; Hatton, 1949;

Krishnamurthi et al., 1960; Reed & McDougal, 1937), except in one

case where shoot growth occurred first (Waynick & Walker, 1930).

This occurs possibly because soil temperatures are lower than air

temperatures in early spring (Krishnamurthi et al., 1960).

Shoot and root growth flushes tend to alternate. During

periods of very active shoot growth the roots are inactive or nearly

so (Crider, 1927; Bevington, 1983; Bevington & Castle, 1985;

Krishnamurthi et al., 1960; Reed & McDougal, 1937). Mendel (1969)

suggests that increases and decreases in the amount of growth

inhibiting substances may govern the cycles. Trunk growth may be

cyclical, coinciding generally with root growth (Krishnamurthi et

al., 1960), or continuous throughout the growing season (Cooper et

al., 1963).









Alternating periods of root and shoot growth and extent of

growth are greatly influenced by the environment. Shoot growth is

reduced below 12.5-13.00 C and above 33-39" C, with optimum growth

occurring between 23-33* C (Mendel, 1969; Reuther, 1973). Root

growth is maximum at soil temperatures of 25-26* C (Reuther, 1973),

and limited below 13 or above 36 C (Castle, 1978). Increases in

soil temperature increase translocation of assimilates to roots

(Vinokur, 1957).

Daylength influences shoot growth of most citrus. While some

citrus is non-responsive to daylength (Warner et al., 1979), many

rootstocks and scions produce more and longer shoots in longer

daylengths (Lenz, 1969; Reuther, 1973; Warner et al., 1979).

The root and shoot relationship is measured many times as a

ratio of fresh or dry weight of shoot and root systems. Mature

citrus trees in California had shoot:root ratios of ca. 3.5

(Cameron, 1939; Cameron & Compton, 1945) and in Florida of 2.2

(Castle, 1978). This ratio varies within a species with age and

environmental or cultural conditions. Some factors that influence

shoot:root ratio of a number of crops are soil moisture levels

(Harris, 1914; Hilgeman & Sharp, 1970; Hsiao & Acevedo, 1974;

Kramer, 1983; Kriedeman & Barrs, 1981; Levy et al., 1983; Menzel et

al., 1986; Rodgers, 1939; Tinus & Owston, 1984), mineral nutrition,

especially nitrogen (Brouwer, 1962; Harris, 1914; Russel, 1977;

Turner, 1922), light intensity (Russell, 1977), and root restriction

(Hameed et al., 1987).


wo









Water Relations

Growth Responses to Water Deficits

Most aspects of plant metabolism are affected by tissue water

relations, but turgor-dependent processes, such as cellular

expansion, are especially sensitive. Water deficits reduce cell

expansion and division, organ growth, or whole plant growth (Hagan

et al., 1959; Hsiao, 1973). Studies with forest trees indicate that

water deficits during bud formation can have as much influence on

shoot length as stress during shoot expansion (Kozlowski, 1983;

Doley, 1981). In addition to reduced production of new leaf area

during periods of water deficits, abscission of older existing

leaves may increase the loss in leaf area (Marsh, 1973; Kriedemann &

Barrs, 1981).

Soil water deficits may result in cessation of growth of a

flush or prolong the period of quiescence between flushes in citrus

(Cooper et al., 1969; Mendel, 1969). Ford (1964) stated that the

alternations between root and shoot growth may become less

pronounced during periods of drought.

Growth reduction in response to water stress may be immediately

followed by an increased growth rate upon relief of stress. This

has been shown with corn (Acevedo et al., 1971; Hsiao et al., 1970),

soybean (Bunce, 1977), tomato (Gates, 1955), pine (Miller, 1965),

barley and rye (Williams & Shapter, 1955) under short term stress.

Similarly, early season moisture stress suppressed dogwood shoot

growth initially, but stimulated late season growth following

regular irrigation (Williams et al., 1987).







8

Root growth is likewise affected by moisture stress, although

to a lesser degree than shoot growth. Reduced meristematic activity

and root elongation as well as reduced water and nutrient uptake due

to suberization occur in response to water deficits in most crops

(Slatyer, 1969). Working with wheat, Cole and Alston (1974) state

that root volume or weight are generally more affected by water

deficits than root length. The rate of suberization may exceed that

of elongation during periods of water deficits, substantially

reducing the non-suberized root area (Kaufmann, 1968; Kramer, 1950;

Turner & Kramer, 1980). Periods of water deficits have reduced

citrus root conductivity (Levy et al., 1983; Ramos & Kaufmann, 1979)

and root elongation (Bevington, 1983; Bevington & Castle, 1985;

Marsh, 1973). Water deficits may result in development of long

roots with restricted branching in citrus (Castle, 1978).

Resumption of root growth following rain or irrigation typically

occurs with most crops (Kozlowski, 1983), however, pine roots

reportedly matured toward the tip during water deficits and became

inactive (Kaufmann, 1968; Lesham, 1965).

Total root growth and distribution of citrus trees are often

altered by irrigation practices. Infrequent irrigation increased

root density or the proportion of total roots deep in the soil

profile when compared to frequent irrigation (Cahoon et al., 1961;

Cahoon et al., 1964; Hilgeman & Sharp, 1970; Marsh, 1973; Ruggiero &

Andiloro, 1985). Partial root zone irrigation in an arid

environment results in an increase in root growth in the wetted soil

volume (Bielorai, 1977, 1982; Bielorai et al., 1981; Cohen et al.,

1987; Rodney et al., 1977). Conversely, high root concentrations







9

within the wetted areas do not occur in regions of abundant rainfall

(Koo, 1980).

Gas Exchange Responses to Water Deficits

Water influences plant growth and development through its

effect on biochemical and physiological processes, including gas

exchange. It is clearly established that stomatal conductance

generally declines under conditions of water deficit. This stomatal

response has been considered to be a feedback response to leaf water

status, involving a reduction of leaf water potential, turgor

potential, or relative water content to a critical point, at which

time the stomata begin to close (Begg & Turner, 1976; Hsiao, 1973;

Raschke, 1975; Turner, 1979). However, correlations of bulk leaf

water status and stomatal conductance are typically weak, and the

relationship is not consistent, but is dependent on developmental

conditions. Numerous recent reports with a variety of crops show

that decreases in stomatal conductance resulting from soil moisture

depletion are not closely correlated with changes in bulk leaf water

status (Bates & Hall, 1981; Bennett et al., 1987; Black et al.,

1985; Blackman & Davies, 1985b; Cock et al., 1985; Gollan et al.,

1985; Lorenzo-Minguez et al., 1985; Osonubi, 1985; Turner et al.,

1985). Furthermore, stomata respond directly to epidermal water

relations and are only indirectly related to bulk leaf water

relations (Edwards & Meidner, 1978). The pressure probe has

recently been used to directly measure turgor of individual cells

(Shackel & Brinckmann, 1985; Frensch & Schulze, 1987) and it is now

evident that large cell turgor gradients may exist between mesophyll







10

and epidermis, illustrating the problems associated with relating

stomatal conductance with bulk leaf water status.

Stronger correlations of stomatal conductance to soil moisture

status than to leaf water status (Bates & Hall, 1981, 1982; Blackman

& Davies 1985b; Gollan et al., 1985; Gollan et al., 1986; Jones et

al., 1983; Osonubi, 1985; Turner et al., 1985) suggest that soil

water deficits may influence stomata in a manner independent of

shoot water relations. The most clear evidence of this comes from

stomatal closure in response to soil water deficits, in spite of

maintenance of leaf turgor by partial root zone irrigation (Blackman

& Davies, 1985 a,b) or by pressurizing the root system (Gollan et

al., 1986). In these cases, stomata responded directly to a signal

from roots under soil water deficits. An interruption in cytokinin

supply has been proposed as this message (Blackman & Davies 1985

a,b; Davies et al., 1986). It has been known for two decades (Itai

& Vaadia, 1965) that soil water deficits affect concentrations of

growth substances in roots and shoots. The well-known involvement

of abscisic acid (ABA) in stomatal action (Davies et al., 1981) and

the interaction of this ABA effect with other phytohormones

(Blackman & Davies, 1985 a,b; Cox et al., 1985; Snaith & Mansfield,

1982) lend support to the suggestions that plant growth substances

play a role in the stomatal response to soil water deficits.

However, stomatal closure of some plants is not related to leaf ABA

levels (Davies & Lakso, 1978; Raschke et al., 1976).

The discovery that stomata respond directly to humidity (Lange

et al., 1971) in a feed-forward manner is well-documented in many

plants. Feedforward responses are important in enabling a plant to









restrict excessive water loss before developing a severe water

deficit. Sharkey (1984) has shown that high transpiration may

result in reduced photosynthetic capacity in a manner similar to

soil water deficits, and suggests that stomatal responses to

humidity may guard against this reduction in photosynthetic

capacity. Stomatal responses to humidity can greatly confuse

interpretation of stomatal responses to soil water deficits. This

stomatal sensitivity to humidity or vapor pressure deficit (VPD) is

well-documented in citrus (Camacho-B., 1977; Camacho-B. et al.,

1974; Fereres et al., 1979; Hall et al., 1975; Kaufmann, 1977;

Kaufmann & Levy, 1976; Khairi & Hall, 1976b; Kriedemann & Barrs,

1981; Levy, 1980; Syvertsen, 1982) and is important in maintaining

favorable leaf water relations under arid environments. Well-

watered citrus trees in environments of drastically different

evaporative demand have been shown to have nearly equal

transpiration and equal or increased leaf water potential in arid

environments due to VPD-induced partial stomatal closure (Kaufmann,

1977; Levy, 1980; Levy & Syvertsen, 1981). Rilgeman (1966) reported

nearly equal transpiration but greater leaf water deficits of citrus

in Arizona compared to Florida. Annual evapotranspiration values

for mature citrus in different environments are quite similar,

ranging from 1.07 to 1.21 m in Florida (Koo, 1963; Rogers et al.,

1983; Rogers & Bartholic, 1975; Reitz et al., 1978; Gerber et al.,

1973), 0.91 to 1.14 m in Texas (Weigand & Swanson, 1982a), and 0.80

to 1.47 m in Arizona (Hilgeman & Sharp, 1970; Hilgeman et al., 1969;

Hoffman et al., 1982).









The literature concerning soil water deficit effects on

photosynthesis is extensive and has been reviewed by many [e.g.

Boyer (1976), Hsiao (1973), Lawlor (1979), Slatyer (1967)]. Early

research suggested that stomatal conductance and CO2 assimilation

are closely correlated for some species, and led to conclusions that

CO2 assimilation was reduced by soil water deficits through stomatal

closure. However, water stress also impairs non-stomatal factors

which may be causal, yet closely correlated with stomatal

conductance. The correlation could result from a concommitant yet

independent effect of water deficits on stomatal and non-stomatal

processes, or from a direct reduction in mesophyll photosynthetic

capacity, followed by stomatal closure in response to reduced CO2

assimilation (Farquhar & Sharkey, 1982; Redshaw & Meidner, 1972;

Wong et al., 1979).

Non-stomatal limitations on photosynthesis are well-documented.

Residual conductance to CO2 is reduced by water stress in many

plants (Brown & Simmons, 1979; Bunce, 1977; Collatz, 1977; Mederski

et al., 1975; O'Toole et al., 1976; Pearcy, 1983; Pellegrino et al.,

1987; Radin & Ackerson, 1981). Stress-induced reduction of

photochemical activity has been shown by using isolated chloroplasts

and artificial electron acceptors or with measurements on

fluorescence (Boyer, 1976; Boyer & Bowen, 1970; Genty et al., 1987;

Keck & Boyer, 1974; Nir & Poljakoff-Mayber, 1967; Sharkey & Badger,

1982; Von Caemmerer & Farquhar, 1981). Reduced carboxylation

capacity in response to stress has also been reported due to

decreases in activity of ribulose bisphosphate carboxylase-

oxygenase, carbonic anhydrase, phosphoenol pyruvate carboxylase,







13

fructose-1,6-bisphosphatase, or sedoheptulose-1,7-bisphosphatase

(Berkowitz & Gibbs, 1982; Boag & Portis, 1984; Huffaker et al.,

1970; Jones, 1973; O'Toole et al., 1976). In situations where

intercellular CO2 concentration is unchanged or higher in stressed

than in control plants, reduced stomatal conductance of stressed

plants is of little importance in photosynthetic reduction (Briggs

et al., 1986; Farquhar et al., 1980; Radin & Ackerson, 1981; Wong et

al., 1979).

Inhibition of leaf growth by water deficits may contribute to a

whole plant reductions in photosynthesis. Water deficits may reduce

the rate of leaf initiation (Husain & Aspinall, 1970) as well as

foliar expansion (Hagan et al., 1959; Hsiao, 1973). Additional loss

of leaf area through abscission of existing leaves may also result

from water deficits. Stress-induced foliar abscission in citrus

normally occurs at a secondary abscission zone between the leaf

blade and petiole (Schneider, 1968). These effects may be quite

damaging, since recovery does not occur until new growth replaces

the lost photosynthetic surface (Hsiao, 1973).

Effects of water deficits on photosynthesis are certainly not

fully understood. The fact that such diverse mechanisms of action

are discussed in the literature is not surprising and all mechanisms

could be applicable under certain situations.

Water deficit effects on gas exchange of citrus has received

fairly limited attention. Hilgeman (1977) stated stomatal closure

of mature citrus in Arizona occurs at progressively earlier times in

the day as soil water is depleted over several weeks. Irrigated and

nonirrigated mature sweet orange trees in Italy began days with









similar stomatal conductances, but conductance of nonirrigated trees

declined steadily throughout the day, while that of irrigated trees

increased to peak around midday before declining (Ruggiero &

Andiloro, 1985). Similarly, leaf conductance of severely-stressed

mature sweet orange trees increased slightly until mid-morning

before declining steadily throughout the day, while conductance of

irrigated trees increased rapidly until mid-morning and did not

decline any until mid-afternoon (Cohen & Cohen, 1983). Hilgeman et

al. (1969) reported increased transpiration of mature sweet orange

trees in Arizona at a soil matric potential of -29 kPa compared to

-79 kPa, especially during the middle part of the day. Koo (1953)

reported citrus tree transpiration in Florida was unchanged at soil

moisture levels between field capacity and depletion of two-thirds

of the readily available water. A reduction in transpiration or

stomatal conductance by soil water deficits has been reported under

many conditions (Bielorai & Mendel, 1969; Brakke et al., 1986; Cohen

& Cohen, 1983; Hilgeman, 1977; Kaufmann & Levy, 1976; Koo, 1953;

Kriedeman, 1971; Ruggiero & Andiloro, 1985; Thompson et al., 1968;

Zekri, 1984). Only controlled conditions utilizing containerized

plants have been used to illustrate a reduction of citrus

photosynthesis by soil water deficits (Bielorai & Mendel, 1969;

Brakke et al., 1986; Kriedemann, 1968, 1971; Ono & Hirose, 1984;

Thompson et al., 1968). Bielorai & Mendel (1969) reported that

sweet lime and rough lemon photosynthesis was reduced more than

transpiration with soil water deficits, while Kriedemann (1971)

reported a greater reduction of transpiration with sweet orange.









Irrigation of Mature Citrus

Irrigation in Florida

Although Florida's climate is characterized by an average

annual rainfall of 120-150 cm, the rainfall is distributed in an

irregular pattern. Consequently, periods of minimal and infrequent

rainfall result in unfavorable soil moisture conditions for optimum

citrus growth and production. Periods of drought combined with the

low water-holding characteristics of most Florida soils create

favorable conditions for the use of irrigation to obtain maximum

production.

Many early studies presented differing results concerning the

need to irrigate mature citrus trees (DeBusk, 1928, 1933; Hoard,

1908; Koo & Sites, 1955; Savage, 1951, 1952; Sites et al., 1951;

Staebner, 1919; Stanley, 1913, 1914, 1916; Stevens, 1909; Williams,

1909; Young, 1948; Ziegler, 1955b). Many of these reports were

based on observations and correlations between rainfall or soil

moisture content and tree response, while some were actual

irrigation experiments. Responses to irrigation and opinions on its

cost effectiveness varied widely. Ziegler (1955b) stated that

irrigation was needed in Florida only in years of light spring

rainfall, and was not essential every year, and suggested the use of

temporary wilt or "anticipated tree wilt" combined with records of

rainfall for practical irrigation timing (Ziegler, 1955a).

Results of a 3-year study were published in 1963 indicating

that irrigation based on soil moisture depletion increased yields

for a variety of scions on rough lemon rootstock (Koo, 1963).

Recommendations were to irrigate from January to June based on







16

one-third soil moisture depletion, and from July to December based

on two-thirds depletion. Since that time irrigation has ranked high

among requirements for maximum production of Florida citrus, even in

years of high rainfall. Thus, irrigation has been used in Florida

to increase yields of many cultivars on different rootstocks (Koo,

1963, 1969, 1975, 1979, 1985; Koo & Hurner, 1969; Koo & McCornack,

1965; Koo & Sites, 1955; Koo et al., 1974; Reese & Koo, 1976).

Likewise, irrigation increases vegetative growth of mature citrus

trees under Florida conditions (Koo, 1963, 1969, 1979, 1985; Koo &

Burner, 1969; Sites et al., 1951; Zekri, 1984). Increases in citrus

yield due to irrigation may result from an increase in canopy volume

(Koo, 1969; Koo & Burner, 1969; Levy et al., 1978). Recommendations

similar to Koo's (1963) concerning more frequent irrigation during

the spring than other parts of the year are reported for other

citrus growing regions (Hilgeman, 1956; Bilgeman & Sharp, 1970;

Mantell, 1977; Weigand & Swanson, 1982a).

Micro Irrigation of Mature Citrus

Micro irrigation systems, which direct water to only a portion

of the soil volume, have received widespread use in citrus culture

around the world. Yields of navel oranges were comparable using

trickle, basin, and dragline sprinklers in South Africa, with ca.

50X savings of applied water for the trickle systems over the other

systems (Bester et al., 1974). Similarly, water savings of 30Z were

possible while using trickle instead of flood irrigation on

grapefruit in Arizona (Roth et al., 1981). Irrigating oranges with

trickle or basin systems in Arizona reduced water application by

greater than 90% when compared with border or high volume sprinkler







17

methods (Roth et al., 1974). Increased tree growth and reduced weed

control problems also resulted from the use of trickle and basin

irrigation. Drip irrigation increased yields and fruit size of

sweet oranges by 30% over flood irrigation in Spain (Legaz et al.,

1981). Similar comparisons of trickle systems and conventional

systems with mature citrus are numerous (Alijbury et al., 1974; Cole

& Till, 1974; Leyden, 1975b; Rathwell & Leyden, 1976; Yagev &

Choresh, 1974) and include those cited in the next discussion.

Irrigation studies frequently use several types of systems or

emitters to apply water to variable percentages of soil surface.

Bielorai (1977, 1978, 1982) reported similar grapefruit yields and

tree growth resulted in Israel from wetting 30-40 as compared to 70%

of the soil surface. Similarly, irrigation of 35, 70, and 90% of

the soil surface for oranges resulted in increased or comparable

yields with the smaller wetted areas (Bielorai et al., 1981). These

results involved data from only the first 3 years following a

reduction in wetted soil volume. Following this period, however,

yields of the 90% wetting treatment were greater than for the other

treatments (Cohen et al., 1985). Another comparison of orange

performance in Israel showed a reduction in growth rate and yield

with partial compared to full wetting (Moreshet et al., 1983).

Other responses included reduced water uptake and tree hydraulic

conductance due to reduced wetted soil volume (Cohen et al., 1985,

1987). Koo and Tucker (1974) recommended in Florida that systems be

designed to wet 60% of the citrus root volume. However, irrigating

high-density oranges at 15-25%, 30-40%, and 70-78% of the under-tree

area showed that yield increased with increasing coverage (Koo,







18

1978). Grapefruit irrigated at 14, 64, and 100% coverage of under-

tree area also responded with increased yield and growth with

increasing coverage (Koo, 1985). Irrigation at 19, 88, and 100% of

under-tree area gave similar results (Zekri, 1984). Five to 10%,

and 28 to 51% coverage of the under-tree area increased yield of

oranges 41-44 and 65%, respectively, over nonirrigated controls in

Florida (Smajstrla & Koo, 1984, 1985).

Fertilization

Uptake and Function of Minerals

Plants require adequate supplies of essential elements for

optimum growth and development. These elements affect plant growth

through their involvement in physiological and biochemical

processes. The amount required by plants, form that is absorbed,

and function vary considerably among the essential elements.

Agricultural soils are more commonly deficient in nitrogen (N)

than any other element. Nitrogen is an essential component of amino

acids, chlorophyll, and many other organic materials, and is

involved in many enzymic processes. Various forms of N are absorbed

by roots, but nitrate and ammonium are the major forms.

Preferential absorption of ammonium (Wallace, 1954; Wallace &

Mueller, 1957) or nitrate (Hilgeman, 1941) has been suggested in

citrus. Several studies have provided estimates of the partitioning

of N throughout the citrus plant. Leaves contained 31% of the N in

a mature orange tree in Florida; with fruit, small roots, branches,

large root, and the trunk containing 27, 21, 13, 5, and 3%,

respectively (Smith, 1963). Cameron and Compton (1945) presented

data from California suggesting that leaves contain 40-50%; twigs







19

and shoots, 10%; trunk and branches, 20-30%; and roots, 15-20% of

the total N. In contrast, leaves contained greater than 60% of the

total N in 3.5 year old orange trees (Cameron & Applemen, 1933).

Averaged over the entire season, N content was 1.86% of the total

tree dry weight.

Absorption and function of the remaining essential elements are

covered in detail by Clarkson and Hanson (1980) and Mengel and

Kirkby (1982) and are briefly outlined here. Carbon, hydrogen,

oxygen, and sulfur have metabolic functions similar to N, in that

they are structural components of essential plant compounds and are

involved in enzymic processes. Sulfur is absorbed from the soil as

sulfate, or from the atmosphere as sulfur dioxide. Carbon,

hydrogen, and oxygen are absorbed in gaseous forms from the

atmosphere. Phosphorus, boron, and silicon are involved in

esterification with alcohol groups in plants, and are absorbed as

phosphates, borate or boric acid, and silicate, respectively.

Potassium, sodium, magnesium, calcium, manganese, iron, chlorine,

copper, zinc, and molybdenum are absorbed by plants in the form of

ions or chelates. Their metabolic functions are numerous, and range

from controlling osmotic and electro-potentials and membrane

permeability to enzyme activation.

Fertilization of Mature Citrus in Florida

Plant growth and yield increase with added fertilizer up to a

critical level, after which no significant increase occurs. As this

level of maximum production is approached, the increase in response

for a specific rate increase becomes less and less. Citrus tree

responses to fertilization do not increase indefinitely with







20

increased fertilization rates, nor do trees accumulate excessive

amounts when minerals are present in large quantities (Roo et al.,

1984; Sites et al., 1953; Sites et al., 1961; Smith & Rasmussen,

1961; Stewart et al., 1961; Wallace et al., 1952). Optimum

fertilization rates depend not only on response to increased

fertilizer application, but on fertilizer and fruit prices as well.

Current recommendations for fertilizing mature citrus groves in

Florida (Koo et al., 1984) are based on many years of controlled

experiments. Mature citrus trees require from 100 to 340 kg

N/ha/year to maintain optimum production, however, rates above 225

kg/ha are rarely justifiable in terms of yield increase for most

citrus cultivars.

Many field experiments have been conducted to determine the

optimum number and timing of fertilizer applications and the effect

of elemental sources on mature citrus trees. Results generally

agree that frequency or timing are of far less importance than

fertilizer rate (Calvert & Reitz, 1964; Reitz, 1956; Reuther &

Smith, 1954; Sites et al., 1953, 1961; Smith, 1970). One to two

applications per year are adequate when recommended rates are used.

Nutrient source is also of little consequence in yield responses of

mature citrus trees (Camp, 1943; Leonard et al., 1961; Reitz, 1956;

Reitz & Koo, 1959; Sites, 1949; Sites et al., 1953; Smith, 1970;

Stewart et al., 1961).

Nitrogen Losses

Applied N may be lost from the soil in many ways. Ammonium

salts and urea may be lost by volatilization of ammonia, especially

on calcareous soils (Volk, 1959; Wahhab et al., 1957).







21

Nitrification, the bacterial oxidation of ammonia to nitrate, can

lead to increased losses of N from the root zone by leaching of

nitrate. Reduction of nitrate to volatile forms of N is carried out

by many species of bacteria in the soil. This process

(denitrification) is promoted by high soil moisture and temperatures

(Mengel & Kirkby, 1982).

About a third of the applied N was lost through leaching in a

newly-developed Florida citrus grove (Calvert, 1975; Calvert &

Phung, 1971). An average of 50 kg N/ha may be lost annually through

leaching in Florida groves (Barnette, 1936). Annual leaching losses

from some southern California citrus orchards have been estimated at

about 67 kg N/ha (Bingham et al., 1971), which is equivalent to

about 45% of the applied N. Wallace et al. (1952) estimated a 15%

recovery of applied N by citrus trees in southern California.

Similarly, annual losses of up to 67 kg/ha have been attributed to

volatilizationn (Chapman, 1951; Chapman et al., 1949).

Controlled-Release Fertilizer Sources

Controlled-release fertilizer sources that are N carriers can

be used to substantially reduce losses of applied N (Maynard &

Lorenz, 1979; Oertli, 1980). Some benefits from using

controlled-release fertilizers are reduced losses of applied N

through leaching, denitrification, or volatilization; the

possibility of less frequent applications; and increased efficiency

in use of materials (Allen & Mays, 1974; Oertli, 1980; Terman &

Allen, 1970). Controlled-release fertilizer sources have proven

beneficial with many horticultural crops (reviewed by Maynard &

Lorenz, 1979). Sulfur-coated urea and isobutylidene diurea have







22

increased mature orange tree yields over ammonium nitrate in Florida

(Koo, 1986).

Young Citrus Tree Care

Objectives of a young citrus tree care program center on

obtaining the greatest amount of growth in the shortest amount of

time. There is little concern for marketable yield or fruit quality

as with mature trees.

Irrigation

Information on irrigation of young citrus is limited. In

Arizona, basin and trickle irrigation systems provided greater

growth and lower water use of young orange trees through 5 years in

the field, compared to border-flood and sprinkler systems (Rodney et

al., 1977; Roth et al. 1974). The trickle system was operated

daily, basin and sprinkler were operated weekly, and flood

irrigations were at intervals of at least 2 weeks. Trees in the

trickle treatment received 11%, in the basin treatment, 13%, and in

the sprinkler treatment, 38% as much water as those in the flood

treatment. Leyden (1975a) compared strip watering, ring watering,

and drip irrigation on newly-planted grapefruit trees in Texas.

Tree growth was similar over a 2-year period, with drip requiring

about 18% and ring watering 25% of the water utilized with strip

watering. Strip and ring watering were scheduled at 30% soil water

depletion, and drip irrigation was scheduled on the basis of pan

evaporation. De Barreda et al. (1984) used drip and basin

irrigation in Spain to compare the application of varying amounts of

water at the same frequency. Applications were based on

coefficients of pan evaporation, and results suggested that









coefficients of 0.10-0.15, 0.20, and 0.30 for the first 3 years

maintained adequate growth. They made no comparison of basin and

drip irrigation methods. Split-root containers have been utilized

to simulate the effects of partial root wetting of young citrus

trees in Florida (Brakke et al., 1986). Preliminary results of gas

exchange measurements suggested that growth may be only slightly

reduced by irrigating 50-75% of the root volume compared to 100%.

Field lysimeters equipped with rain shelters have been used to

study irrigation requirements of young orange trees in Florida

(Aribi, 1985; Smajstrla et al., 1985). Irrigations were scheduled

at matric potentials of -10, -20, and -40 kPa, which corresponded to

available soil water depletions of 30, 45, and 55%. Water was

applied to return the upper 60 cm of soil to field capacity, thus

23, 34, and 42 liters were applied per irrigation per tree in the

-10, -20, and -40 kPa treatments, respectively. Maintaining

weed-free conditions around trees resulted in a 50Z reduction in

water use compared to trees with a bahiagrass cover. Tree growth

was greatest when irrigations were scheduled at -20 kPa and the

ground was maintained weed-free.

General recommendations have been given for irrigation of

newly-planted citrus in Florida. Jackson and Ferguson (1984)

recommend that newly-planted trees irrigated by the basin method

should be watered 2 to 3 times per week for 8 weeks, and once per

week thereafter. Ziegler and Wolfe (1975) suggest applying 30-38

liters per tree each 2 weeks in the spring, with no irrigation

needed in the fall. Jackson and Lawrence (1984) recommend a

"generous supply" of water applied every 7-10 days.









Fertilization

Optimum nutrient levels for maximum canopy growth should be

maintained by fertilization of young citrus trees (Koo & Reese,

1971). Young tree fertilization practices in Florida have changed

over the years. Collison (1919) made a 10-year comparison of

fertilizer rates. The standard treatment used 0.136 kg N/tree/year

in year-1, and was gradually increased to 0.408 kg N/tree/year by

year-10. Some plots received one-half, twice, and four times this

amount. After several years, trees in the standard and one-half

standard plots had larger trunk diameters than two-fold and

four-fold treatments. Although not properly replicated, these data

suggest as little as 0.068 kg N/year are needed for fertilization of

young citrus trees. Bryan (1940) recommended applying 0.07 kg

N/year per meter of canopy spread for young citrus trees in Florida.

Several field experiments conducted in the 1950s suggested that

annual rates of 0.073 kg N/tree for the first 2 years were

sufficient to obtain optimum tree growth on the Ridge (Rasmussen &

Smith, 1961, 1962). These rates and the authors' recommended

application frequency of three times during the first year and two

times thereafter differed from the average grower practice. Up to

seven applications per year and rates of 0.13 and 0.25 kg N/tree for

the first 2 years, respectively, were not uncommon at the time.

Calvert (1969) reported that trees responded with greater

growth to rates of 0.22 to 0.32 kg N/tree than 0.11 kg/tree when

growing in the flatwoods in marginal soils. Furthermore, four

applications per year were superior to three. These data illustrate









the importance of considering soil type when evaluating

fertilization needs.

There is a wide range of adequate fertilization methods

available for growers, and considerable judgement is required to

choose the most efficient. Current recommendations (Koo et al.,

1984) call for 0.18-0.22 kg N/tree in year-1 and 0.29-0.36 kg N/tree

in year-2. Other elements should be applied in proportion with N in

the following ratio: N-l, P205-1, K20-1, Mg-1/5, Mn-1/20, Cu-1/40,

B-1/300.

Controlled-release N sources have been used for fertilization

of young citrus trees. Controlled-release isobutylidene diurea

increased growth of young container-grown citrus compared to soluble

N sources (Khalaf, 1980; Khalaf & Koo, 1983). In the same study,

isobutylidene diurea and sulfur coated urea reduced N leaching

losses compared to soluble sources. Fucik (1974) similarly

demonstrated an increase in growth of young, container-grown citrus

trees with controlled-release compared to soluble fertilizer.

Jackson & Davies (1984) reported similar growth rates of young,

field-grown 'Orlando' tangelo trees occurred with sulfur coated urea

and a soluble fertilizer source, but application frequency was

reduced by 50% with sulfur coated urea.

Container- and Field-grown Nursery Trees

Cultural practices and growing conditions of any nursery affect

growth and development of plant material in the nursery, but also

after transplanting to the field. Production systems for field and

greenhouse citrus nurseries are well-developed, however, little

effort has gone into understanding the effect of nursery tree







26

characteristics on growth after field planting. Webber (1932), with

'Washington' navel orange, and Gardner and Horanic (1959) with

'Parson Brown' and 'Valencia' orange reported no relationship

between initial size of nursery trees and mature tree size. Effects

on precosity were not reported. Grimm (1956, 1957) stated the most

important factor affecting initial growth of bare-rooted nursery

trees was the protection of roots while they were out of the ground.

The advantages of containerized, greenhouse nursery systems are

varied and include greater control over the production system, lower

land requirement, and shorter production cycle (Castle et al., 1979;

Moore, 1966; Platt & Opitz, 1973; Richards et al., 1967). Nursery

trees produced under these conditions are much different from the

traditional field-grown nursery trees. Shoot:root ratio of nursery

trees is substantially increased in this system over the field

system (W.S. Castle, Citrus Research and Education Center, Lake

Alfred, personal communication). Very high nitrogen and water

applications, reduced light intensity, and root restriction by

containers may all contribute to this, as all of these factors have

been shown to increase shoot:root ratio. Vigorous root growth is

altered when the available soil volume is permeated, at which time

the growth pattern may be shifted to fibrous roots (Castle, 1978).

This results in a substantial increase in the proportion of fibrous

to non-fibrous roots of container-grown compared to field-grown

citrus nursery trees (W.S. Castle, Citrus Research and Education

Center, Lake Alfred, personal communication).

Initial expansion of the root system is critical for successful

establishment of any containerized transplant (Castle, 1987). The









potential for rapid initial root growth of containerized forest

seedlings has been studied in detail and is directly linked to

survival and tree growth after field planting. Container

characteristics such as size and shape have altered post-planting

shoot and root growth of container-grown transplants (Elam et al.,

1981; Hiatt & Tinus, 1974; Hite, 1974; Tinus & Owston, 1984; Van

Eerden & Arnott, 1974). Container medium has also affected tree

growth following field planting in a number of species (Elam et al.,

1981; Hellum, 1981) including citrus (Warneke et al., 1975). Elam

et al. (1981) reported considerable variation between oak species

with respect to the effects of container and media characteristics

on growth. Less root growth of pine seedlings occurred as the

length of time plants were maintained in containers was lengthened

(Hellum, 1981). Reduced irrigation frequency or nitrogen

fertilization prior to removal from the nursery has been used to

increase root and shoot growth of container-grown trees (Rook, 1973;

Timmis, 1974; Tinus & Owston, 1984).

Drying of container medium after planting in the field may

induce plant water stress in some cases. A substantial increase in

drainage out of the medium occurred following removal from a

container and placement in contact with field soil (Costello & Paul,

1975; Nelms & Spomer, 1983; Warneke et al., 1975). Low survival

rates have been attributed to severe water stress due to these

conditions.

Post-plant growth comparison of container- and field-grown

citrus nursery trees has been conducted in Texas. Leyden and Timmer

(1978) observed growth of grapefruit on sour orange trees for 2.5







28

years in the field and concluded that container-grown trees would be

less productive and smaller than field-grown trees during the early

years of bearing. Maxwell and Rouse (1980, 1984) reported that

container-grown grapefruit on sour orange trees remained smaller

than field-grown trees through 10 years after planting, but yield

did not differ. Container-grown trees were not produced under

greenhouse conditions as is the case in Florida, and field-grown

trees were transplanted as ball and burlapped stock in both studies.
















CHAPTER III


MICROSPRINKLER IRRIGATION AND GROWTH OF YOUNG 'HAMLIN'
ORANGE TREES. I. CANOPY GROWTH AND DEVELOPMENT



Introduction

Comparable or increased growth or yield and decreased water use

by mature citrus has been observed in several countries when micro

irrigation systems were compared with flood or high volume sprinkler

systems (Alijbury et al., 1974; Bester et al., 1974; Bieloral, 1982;

Bielorai et al., 1981; Legaz et al., 1981; Roth et al., 1981). Some

have found that increasing the area of coverage by irrigation

systems has increased growth and yield of mature citrus (Moreshet et

al., 1983; Koo, 1978, 1985; Zekri, 1984). Smajstrla and Koo (1984,

1985) reported an increase in yield over non-irrigated trees in

Florida when irrigating between 5 and 50% of the under-tree canopy

area of mature citrus. Koo and Tucker (1974) recommended that 60%

of a citrus tree's root volume be covered by an irrigation system.

A number of studies have been conducted comparing growth and

development of young citrus trees using various irrigation methods.

Border-flood, sprinkler, basin, and trickle irrigation methods were

compared on young 'Campbell Valencia' orange trees in Arizona

(Rodney et al. 1977; Roth et al. 1974). Tree growth and yield were

greater and water use lower in the first 5 years with the basin and


L







30

trickle systems compared with the other systems. Growth of young

navel orange trees has been evaluated using drip and basin

irrigation methods in Spain (De Barreda et al., 1984). Treatments

consisted of applying different volumes of water at the same

frequency based on coefficients of pan evaporation. Coefficients

were increased with each year of age, suggesting that coefficients

of 0.10-0.15, 0.20, and 0.30 for the first 3 years maintained

adequate growth. Leyden (1975a) compared irrigation systems on

newly-planted 'Star Ruby' grapefruit trees in Texas. Strip and ring

watering were scheduled based on 30% soil water depletion, and drip

irrigation was scheduled on the basis of pan evaporation. Drip

irrigation used 18 and ring watering used 25% of the cumulative

irrigation water required for strip watering. Tree growth was

similar between the three systems over a 2-year period.

Irrigation of young citrus trees in Florida is also necessary

to obtain optimum growth. Irrigation studies on newly-planted

'Valencia' orange trees were conducted using a series of lysimeters

under rain shelters (Aribi, 1985; Smajstrla et al., 1985).

Comparing irrigation scheduling at matric potentials of -10, -20,

and -40 kPa, tree growth was greatest when irrigations were

scheduled at -20 kPa and the ground was maintained weed-free.

Split-root containers were utilized to simulate the effects of

partial root-volume wetting of young 'Hamlin' orange trees (Brakke

et al., 1986). Preliminary results of gas exchange measurements

suggested that growth may be only slightly reduced by irrigating

50-752 of the root volume compared to 100%. Allen et al. (1985)







31

stated that frequent irrigation of containerized citrus rootstock

seedlings was required to sustain high photosynthesis.

An estimated 6-10 million young citrus trees are being planted

in Florida annually (Jackson et al., 1986), however, water

requirements, irrigation scheduling, and growth responses to

microsprinkler irrigation have not been studied under field

conditions. The purpose of this field study using 'Hamlin' orange

trees was to determine 1) the optimum level of soil water depletion

at which irrigations should be scheduled to maximize growth, 2) the

amount of irrigation time needed to replenish soil water to field

capacity, and 3) the effect of irrigation pattern on growth of

second season trees.

Materials and Methods

Site and Soil Characteristics

Three field experiments were conducted at the Horticultural

Research Unit near Gainesville, FL, in 1985, 1986, and 1987. Soil

type was Kanapaha sand (Carlisle et al., 1988) (loamy, siliceous,

hyperthermic, Grossarenic, Paleaquults) underlain by a hardpan.

Particle size distribution was 93.4% sand, 3.9% silt, and 2.7% clay.

The soil has a field capacity of ca. 11.3%; permanent wilting point,

ca. 2%; mean bulk density, 1.56 g/cm 3; pH, 6.4; and percent organic

matter, 0.65. A desorption soil water characteristic curve was

determined using a vacuum desiccator employing undisturbed soil

cores in weighable pressure cells fitted with fritted glass plates

(Fig. 3-1). Saturated hydraulic conductivity was 9.3 cm/hr as

determined by using undisturbed cores in a constant head

permeameter.


--;m- ..














-1500


-20

-10

-5


0.1 0.2 0.3


Volumetric water content


(m3.m3)


Fig. 3-1. The relationship between soil water content
and water potential of Kanapaha sand at the Horticultural
Unit.







33

Beds (16.75 m width x 0.60-0.75 m height x 85 a length) were

constructed in March, 1985 to facilitate drainage and simulate

flatwoods growing conditions. Ground water table was monitored

through observation wells located in the tree rows and averaged 1.1

m deep between June and Sept. (rainy season) with a minimum of 0.45

m, and greater than 1.6 m during other months. A ground cover of

bahiagrass was developed between rows and was mowed as needed, while

tree rows (about 2.1 m width) were maintained weed-free with

herbicides.

Irrigation Treatments

The irrigation system was designed so that each treatment could

be monitored and controlled individually using a flow meter and

pressure gauge controlled by a gate valve. Water was conveyed from

a manifold through 2.54 cm PVC main lines and 1.91 cm black

polyethylene laterals down tree rows. Three tubes, one per

treatment, were positioned 50-55 cm west of each tree row which ran

the full length of the bed, allowing unrestricted randomization at

each experimental site. Irrigation water containing 287 mg/liter

total dissolved solids was applied via 90", 38 liter/hr Maxijet"

microsprinkler emitters positioned 1 m northwest of each tree. This

positioning also provides optimum cold protection for young citrus

trees (Rieger et al., 1986). The area of ground covered was about 5
2
m for these emitters. Irrigation was controlled by hand regulation

of gate valves to maintain the system pressure at about 140 kPa.

Irrigation scheduling was based on soil water content as

monitored by the neutron scattering method (Hillel, 1982) using a

Troxler Model 1255 neutron probe. A calibration curve for the


^.









neutron probe was developed by the gravimetric method (Fig. 3-2)

(Hillel, 1982). Access tubes constructed of 48 mm i.d. aluminum

irrigation tubing were driven into the soil 1 m from the emitters

and 35 cm from each of four randomly chosen trees per treatment in

1985 and 1986 and from each of three trees per treatment in 1987.

Soil moisture measurements were made daily or as needed during the

rainy season at a depth of 30 cm. The soil volume around the tree

was irrigated to field capacity when a prespecified level of

available soil water depletion (SWD) was reached. Irrigations were

initiated when any of the four neutron probe readings reached the

level of SWD in 1985 and 1986, and based on the average of the three

neutron probe readings in 1987. Three levels of irrigation were

designated as high (202 SWD), moderate (45% SWD), and low (652 SWD).

The amount of irrigation water needed for each treatment to

replenish soil moisture to field capacity was determined during the

first few irrigations in 1985. Soil water content at 30 cm depth

was monitored using the neutron scattering method at 15-min

intervals following an irrigation. By initially varying the length

of irrigation time, the approximate length of time needed for each

treatment to return soil moisture from the specified SWD level to

field capacity was determined.

Plant Material and Experimental Design

Commercially obtained, bare-rooted 'Hamlin' orange [Citrus

sinensis (L.) Osb.] on sour orange (C. aurantium L.) trees were

planted on double row beds at 7.6 m between and 4 m within rows.

Initial trunk diameter averaged 1.2, 1.4, and 1.5 cm in 1985, 1986,

and 1987, respectively. A randomized complete block design






















0.6
y 4.29x 0.05

o.s5- r 0.99


S 0.4


0.3


0.2


0.1 I I I I I I
0.04 0.06 0.08 0.1 0.12 0.14 0.16
Volumetric water content (m3*m-s)













Fig. 3-2. The relationship between neutron probe ratio and
volumetric water content in Kanapaha sand at the
Horticultural Unit. R counts per min. in soil,
R = standard counts per min.
o







36

consisting of four blocks was utilized in 1985 and 1986, employing

six single-tree replications per treatment per block, resulting in a

total of 24 replications per treatment. A completely randomized

design with 13 replications was utilized in 1987 because the block

effect was not significant in 1985 and 1986. Trees were planted on

10-12 May, 3-4 May, and 2 April for the 1985, 1986, and 1987

experimental periods, respectively. All trees were irrigated every

2 days during an establishment period of 10-14 days in each

experiment. Commercial fertilization rates recommended for Florida

(Koo et al., 1984) were followed using an 8N 2.6P 6.6K 2Mg -

0.2Mn 0.12Cu 0.2Zn 1.78Fe dry formulation. Four applications

were made at regular intervals in 1985, while five were made in 1986

and 1987.

Plant Measurements

Seasonal measurements. Shoot growth was measured throughout

the growing season during 1985 and 1986. As shoots began growth,

three per tree were randomly chosen and tagged. Shoot length was

measured with a ruler at about 7-day intervals thereafter. The

length and width of the third leaf from the shoot base was measured

at the same time. This leaf position was used because the bottom

two leaves were many times of unrepresentative size or shape. Leaf

expansion was determined by using the product (x) of length and
2
width in an equation for leaf area (area 2.33 + 0.63x, r = 0.94).

This equation was derived from a sample of 100 leaves of various

sizes by linear regression of leaf area, determined with a LI-COR

Model LI-3000 leaf area meter, on leaf length x width (data not

shown). Final shoot length and leaf area were compared, as were


i .---.









expansion rates. Rates were calculated on a daily basis by

subtracting the initial length or area when shoots were tagged from

the final length or area, and dividing by the number of days

required to reach the final size. Although shoot and leaf expansion

were not measured in 1987, dates of initial shoot growth were

recorded for each tree in each of the three flushes. Intervals of

about 7 days were again used. The percentage of the tree population

growing for each irrigation treatment and at each date was

calculated for all 3 years.

Canopy and trunk measurements. Canopy height and width in two

directions were measured initially and in December of each year (15

May and 10 Dec. 1985, 6 May and 7 Dec. 1986, 4 April and 18 Dec.

1987). Width measurements were averaged and canopy volume was

calculated as (4/3)(3.14)(1/2H)(1/2W)2, where H-height and W-width

(Westwood, 1978). This formula most closely approximates the canopy

shape of a young tree which is taller than it is wide. A mark was

painted with latex paint about 5 cm above the bud union where trunk

diameter was measured in two directions with a hand-held caliper.

Measurements were made on the same days as canopy measurements. The

two measurements were averaged and trunk cross sectional area

calculated. In addition total shoot length for each tree was

measured with a ruler within 1 week of planting dates.

Final measurements. Root excavation of 20, 21, and five plants

per treatment in 1985, 1986, and 1987, respectively, was conducted

by hand in Dec. of each year. Initially a circular trench was dug

about 40 cm deep at a distance of 120 cm from the trunks. The few

roots extending beyond this distance were individually recovered by







38

excavation. The entire shallow root system was recovered almost

completely intact by undercutting it to a depth of 40 cm until the

sand loosened exposing the roots. The depth was increased near the

trunk to ensure recovery of the taproots. This operation was not

difficult due to the loose, sandy nature of the soil and shallowness

of the root systems. Trees were taken indoors where roots and

canopies were separated. Root growth and distribution are discussed

in Chapter IV.

Leaves were removed from canopies for fresh weight

determination. Leaf area was calculated from the fresh weight

utilizing a linear relationship obtained from sampling eight trees.

Leaf area was measured on these trees using a LI-COR Hodel LI-3000

leaf area meter. The relationship shown in Fig. 3-3 was derived

from linear regression of leaf area on fresh weight. Total shoot

length was measured with a ruler following leaf removal.

A random sample of four, six, and five trees in 1985, 1986, and

1987, respectively, were separated into the three growth flushes

that had occurred during the experimental period. Shoots within

each flush were counted and measured with a ruler. Mean and total

shoot length and mean shoot number were calculated for each growth

flush.

All canopy parts were dried in an oven at 80 C to determine

canopy dry weight.

Analysis of Data

Shoot length, leaf area, and expansion rates measured

throughout the growing season were analyzed separately for each of

the three flushes in 1985 and 1986. Shoots in each flush were


_ _























2.6 y = .00416x 0.53

2.2 r2 0.93
CI 2.2-
E
1.8 -


M 1.4
(D

1

300 400 500 600 700
Fresh weight (g)













Fig. 3-3. Relationship of leaf area and fresh weight from
eight representative 'Hamlin' orange trees after 8 months
in the field.


1







40

blocked into groups of shoots that had begun growth within periods

of time not exceeding 2 weeks in length. Data were analyzed by

analysis of variance as a treatment x time period factorial

experiment.

Data concerning the percentage of each treatment population

growing throughout the season were analyzed using linear, quadratic,

and cubic regression models. Models with the highest level of

significance and best fit for each treatment population were chosen.

Resulting equations were tested by analysis of covariance for

homogeneity within each growth flush.

Shoot number, mean length, and total length measured by

dividing excavated trees into different flushes were analyzed by a

split-plot analysis with treatments as main plots and flushes as

subplots.

Final plant measurements from different years were analyzed

separately. Data on dry weight and leaf area were subjected to

analysis of variance and data on canopy volume, trunk cross

sectional area, and shoot length to analysis of covariance to

standardize differences in initial plant measurements. Where

irrigation levels were significantly different, Williams' (Williams,

1971) test was used to compare means. This test is useful in cases

where curve-fitting is not desired or is difficult due to a small

number of dose levels. It was employed due to the small number of

treatments to compare irrigation levels to the most frequent

scheduling treatment to determine the level at which a significant

response occurred.









Microsprinkler Irrigation Spray Patterns

Two experiments were conducted on the same site to test the

effect of microsprinkler irrigation spray pattern on 'Hamlin' orange

tree growth the second season in the field. Site characteristics

were as described for microsprinkler irrigation scheduling

experiments, except trees were set 3.4 m apart in the rows. Two

microsprinkler spray patterns were utilized, 90 and 180.

Approximately 89 liters were applied per tree at each irrigation,

delivered through 89 liter/hr emitters and distributed over ca. 5.9

and 9.3 a2 of ground area for the 90 and 180 patterns,

respectively.

Treatments begun on 10 May 1986 in the first experiment on

trees planted in May 1985. The factorial experiment performed from

May 1985 May 1986 was used to study two types of nursery trees

receiving four types of fertilizer (Chapters VI & VII). Each 2 x 4

treatment combination had been assigned to two trees within each of

four randomized complete blocks in 1985. Subsequently, each spray

pattern was randomly assigned to one of the two trees of each tree

type fertilizer type combination in May 1986. Treatments were

begun on 2 May 1987 in the second experiment on trees planted in May

1986 (Chapters VI & VII). Each 2 x 4 treatment combination had been

assigned to two trees within each of two randomized complete blocks

in 1986. Each spray pattern was again randomly assigned to one of

the two trees of each tree type fertilizer type combination in May

1987.

Trees were grown during the first year under 90 microsprinkler

irrigation on 20% SWD scheduling. During the two experimental


_ )__









periods irrigations were scheduled when 20% SWD was reached in

accompanying irrigation scheduling experiments.

Trunk cross sectional area and canopy volume were measured as

described previously. Measurements were made on 9 May and 7 Dec.

1986 for the first experiment and 30 April and 18 Dec. 1987 for the

second. Measurements were analyzed separately for the two

experimental periods. Split-plot analysis was utilized with

fertilizer type x tree type as main plots and irrigation patterns as

subplots. Analysis of covariance was used to standardize

differences in plant measurements from the beginning of the

experimental periods.

Environmental Variables

Rainfall was recorded daily with a Science Associates Model 503

rain gauge located about 200 m east of the experimental site.

Relative humidity and temperature were recorded continuously using a

hygrothermograph (WEATHERtronics Model 5021) located at the

experimental site.

Results and Discussion

Irrigation Amount and Frequency

Thirty-eight, 50, and 76 liters/tree were needed for the high,

moderate, and low irrigation treatments, respectively, to return the

soil to field capacity (Table 3-1). These durations resulted from

measurements of soil moisture content by the neutron probe at a

depth of 30 cm, and did not take into account the level of depletion

in the surface soil. Irrigation durations of 1-2 hr were used with

the 38 liter/hr, 90 emitters used in this study. An industry

survey has shown that many times growers do not take advantage of





















Table 3-1. Duration of irrigation and amount of water
applied at each irrigation as related to soil water
depletion for young 'Hamlin' orange trees.



Soil water Irrigation Amount applied

depletion duration liters/

(Z)Z (hr) tree amy



20 (Highx) 1.0 38 7.5

45 (Mod.) 1.3 50 10.0

65 (Low) 2.0 76 15.1



Based on neutron probe measurements at 30 cm depth.
YBased on area wetted by emitters of about 5 m .
xHigh, moderate, and low refers to irrigation frequency.







44

the water savings that are possible with micro irrigation systems

due to excessive operating time (Hutcheson & Bellizio, 1974).

The number of irrigations needed to maintain soil moisture at

the specified levels differed greatly (Table 3-2). An average of

31, 11, and 2 irrigations per season were required for the high,

moderate, and low schedules, respectively. During dry periods,

irrigations were 2-3 days apart in the high treatment and 4-6 days

apart in the moderate schedule. The number of irrigations in the

moderate schedule varied from year-to-year more than the other

scheduling treatments. The 16 irrigations needed in 1986 were

nearly twice the number needed in the other years. Different

factors were probably responsible for the reduced number of

irrigations needed in 1985 and 1987 compared with 1986. In 1985,

more frequent rainfall (Fig. 3-4) did not allow SWD to reach 45%

(moderate) as often as occurred in 1986 (Fig. 3-5). Rainfall was

not responsible for the reduced number of irrigations in 1987 since

frequency and cumulative amount of rainfall was lower (Fig. 3-6)

than in 1986. Tree growth and water use was less during the 1987

season regardless of irrigation treatment, thus the plants required

fewer irrigations. Over the 3-year period, trees in the moderate

and low treatments received respectively 49 and 13% as much

irrigation water as those in the high treatment (Table 3-2).

Individual Shoot and Leaf Growth Seasonal Measurements

During the first flush of 1985 trees under the high and

moderate treatments made significantly greater average shoot growth

(P<.0165) and leaf growth (P<.0315) than trees under the low

treatment (Table 3-3). Shoot and leaf size in the second and third















I I I I I I I














I I I






I I I I I I I


J


A
Time (months)


N 0


Fig. 3-4. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, May-Dec. 1985. High, moderate, and
low refer to irrigation treatments based on 20, 45, and 65% soil
water depletion, respectively.


M J













Table 3-2. Number of irrigations and cumulative water applied
for young 'Hamlin' orange trees under scheduling treatments
based on soil water depletion.



Soil Cumulative

water Irrigations/ Cumulative water applied irrigation

depletion year liters/ and rain

(%)z (no.) tree (cm)y (cm)



1985

20 (Highx) 31 1173.5 23.3 141.2

45 (Mod.) 8 402.8 8.0 125.9

65 (Low) 1 75.7 1.5 119.4

1986

20 (High) 35 1324.9 26.4 145.8

45 (Mod.) 16 805.5 16.0 135.5

65 (Low) 4 302.8 6.0 125.5

1987

20 (High) 27 1022.0 20.3 112.1

45 (Mod.) 10 503.5 10.0 101.8

65 (Low) 1 75.7 1.5 93.3


ZBased on neutron probe measurements at 30 cm depth.
'Based on area wetted by emitters of about 5 m .
xHigh, moderate, and low refers to irrigation frequency.























S I I II
2







.2
S2





2-

0
-I
-* 1 2

E
i12 ---r1 1111-- i ---- 1 -
0


C



M J J A 8 0 N D
Time (months)



Fig. 3-5. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, May-Dec. 1986. High, moderate, and
low refer to irrigation treatments based on 20, 45, and 65% soil
water depletion, respectively.















*I 1II I I I I
E
2-


21-









I I I I I I
E
2
0

2

0


E

=8C




I I I I I I I
A M J A S 0 N D

Time (months)





Fig. 3-6. Distribution of rainfall and microsprinkler irrigation
at the Horticultural Unit, April-Dec. 1987. High, moderate, and
low refer to irrigation treatments based on 20, 45, and 65% soil
water depletion, respectively.


__









Table 3-3. Shoot and leaf sizes and expansion rates for three
growth flushes of young 'Hamlin' orange trees as influenced by
irrigation based on soil water depletion, 1985.



Soil water Shoot Leaf

depletion length rate area rate

(%) n (cm) (cm/day) n (cm2) (cm 2/day)



Flush one (5/30 6/25)z

20 (High) 70 9.7 0.5 67 22.6 0.8

45 (Mod.) 69 10.8 0.6 67 23.3 0.9

65 (Low) 64 8.6 0.4 61 19.6 0.7

SEy 0.7 0.1 1.4 0.1

Flush two (7/12 8/19)

20 (High) 49 21.5 1.2 49 38.7 1.4

45 (Mod.) 40 22.3 1.3 40 40.9 1.6

65 (Low) 34 17.4 1.0 33 28.1 1.0

SE ns ns ns ns

Flush three (9/11 11/25)

20 (High) 36 28.8 1.1 37 44.1 1.7

45 (Mod.) 28 31.5 1.2 28 43.8 1.6

65 (Low) 10 25.8 1.2 10 40.7 1.5

SE ns ns ns ns



ZRange in dates of growth initiation.
YSE standard error of mean.







50

flush were not dependent on irrigation treatments, although the low

treatment consistently had the lowest means.

Irrigation treatment effects on growth of the first flush in

1986 were in contrast to 1985, in that trees in the moderate and low.

treatments had greater individual shoot length (P<.0077) and leaf

area (P<.0005) than trees in the high treatment (Table 3-4). A

different pattern occurred in flush two, when shoot length of trees

in the high and moderate treatments was larger (P<.0117) than those

in the low treatment. Leaf size in flush two and both shoot and

leaf size in flush three were independent of irrigation treatments.

Shoot length, leaf area, and rates of growth generally increased

from flush one to flush three in both 1985 and 1986.

Shoot and leaf expansion rates averaged less in flush one than

the other flushes. Similarly, growth rates of the spring flush of

mature orange trees in Florida, estimated as gain in biomass, were

less than the summer flush (Syvertsen et al., 1981). Shoot growth

rates ranged from 0.4-1.0 cm/day in flush one and 0.7-1.4 cm/day in

the other flushes, which was considerably more than previously

reported for mature trees of various citrus species in subtropical

India (Krishnamurthi et al., 1960).

Shoot Number and Length Within Flushes Final Measurements

Shoot number, average shoot length, and total shoot length were

not significantly affected by irrigation treatments in 1985.

Cumulative shoot length during the season (three flushes summed) was

650, 760, and 537 cm/tree in the high, moderate, and low treatments,

respectively. Lack of significant differences may have been be due

to the small number of replications (four) or to the loss in









Table 3-4. Shoot and leaf sizes and expansion rates for three
growth flushes of young 'Hamlin' orange trees as influenced by
irrigation based on soil water depletion, 1986.



Soil water Shoot Leaf

depletion length rate area rate

(%) n (cm) (cm/day) n (cm2) (cm2/day)



Flush one (5/25 6/17)z

20 (High) 65 13.7 0.8 65 15.9 0.6

45 (Mod.) 45 17.5 1.0 44 22.0 0.9

65 (Low) 49 17.8 1.0 49 21.4 0.7

SEy 1.5 0.1 1.9 0.1

Flush two (7/7 9/13)

20 (High) 61 20.3 1.1 60 31.3 1.1

45 (Mod.) 39 21.6 1.1 44 34.9 1.3

65 (Low) 40 17.4 0.7 40 30.4 1.1

SE 1.5 0.1 ns ns

Flush three (8/30 12/3)

20 (High) 58 30.6 1.4 53 46.6 2.1

45 (Mod.) 49 31.7 1.4 46 46.3 1.8

65 (Low) 45 28.2 1.3 42 37.3 1.5

SE ns ns ns ns



ZRange in dates of growth initiation.

YSE standard error of mean.


_j


''







52

precision in the whole plot analysis of treatments in the split-plot

design. Across all treatments, however, individual shoot length

significantly increased (P<.0001) and shoot number per flush

decreased (P<.0108) successively from flush one to flush three

(Table 3-5). Flushes two and three added significantly (P<.0124)

more total shoot length than flush one during the season (Table

3-5).

In contrast, irrigation treatments significantly affected shoot

number per flush (P<.0079) and total shoot length per flush

(P<.0164) in 1986, although there was no effect on mean shoot length

(Table 3-6). A mean of 19 shoots per flush occurred on trees in the

high treatment, with 13 and 9 shoots per flush occurring on trees in

moderate and low treatments, respectively (SE-3). Total length per

flush was 335 cm for trees in the 20% SWD treatment, and 258 and 134

cm for trees in the 45 and 65% SWD treatments (SE-64). Reduction in

total length per flush by 65% SWD was more pronounced in flush three

(Table 3-6). Cumulative shoot length for the entire season was 1004

cm in high, and 774 and 402 cm in moderate and low treatments,

respectively. As in 1985, mean shoot length, pooled over the

treatments, significantly (P<.0001) increased from flush one to

flush three, while flush three had significantly (P<.0012) fewer

shoots than flushes one and two (Table 3-6). These trends balanced

total length of individual flushes such that the flushes were not

different.

Irrigation treatments did not affect mean shoot length or

number in 1987, but significantly (P<.0204) altered total shoot

length per flush, producing a total length of 408, 235, and 212 cm









Table 3-5. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin' orange trees
as related to irrigation based on soil water depletion, 1985.



Soil water Shoot Total

depletion Shoot length length

(Z) no.z (cm)z (cm)z



Flush one (5/30 6/25)y

20 (High) 17 9.5 163.0

45 (Mod.) 15 7.9 118.7

65 (Low) 18 7.9 146.4

Mean 17 8.4 142.7

Flush two (7/12 8/19)

20 (High) 13 20.5 261.3

45 (Mod.) 10 28.2 289.1

65 (Low) 8 25.7 211.7

Mean 10 24.8 256.4

Flush three (9/11 11/25)

20 (High) 7 34.8 225.9

45 (Mod.) 11 32.2 352.4

65 (Low) 4 42.9 179.1

Mean 7 36.6 262.1

SEx 2 3.4 36.9



ZMeans of 4 trees/treatment.
YRange in dates of initiation.
xSE for comparison of means among growth flushes. Comparison
among irrigation levels within growth flushes is inappropriate
since irrigation x flush interaction is not significant.


_ _









Table 3-6. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin' orange trees
as related to irrigation based on soil water depletion, 1986.



Soil water Shoot Total

depletion Shoot length length

(%) no. (cm)z (cm)z



Flush one (5/25 6/17)y


20 (High)

45 (Mod.)

65 (Low)

Mean


13.1

13.0

10.8

12.3


284.3

203.7

126.3

201.9


Flush two (7/7 9/13)


20 (High)

45 (Mod.)

65 (Low)

Mean


15.9

18.2

20.1

18.1


366.4

283.5

234.1

301.1


Flush three (8/30 12/3)


20 (High)

45 (Mod.)

65 (Low)


Mean

SEx


29.5

33.4

22.9

29.5


353.6

286.7


41.7


219.9


ZMeans of 6 trees/treatment.
Range in dates of initiation.
xSE for comparison of means among growth flushes. Comparison
among irrigation levels within growth flushes is inappropriate
since irrigation x flush interaction is not significant.









of growth per flush (SE=59) in the high, moderate, and low

treatments, respectively (Table 3-7). As in 1986, the treatment

effect was most pronounced in flush three. Mean shoot length

significantly increased (P<.0001) and shoot number decreased

(P<.0001) successively from flush one to flush three, although total

length was not different among the three flushes. Cumulative shoot

length for the season averaged 1225 cm for the high, and 706 and 635

cm for the moderate and low treatments.

Severe defoliation shortly after transplanting trees from the

nursery to the field in 1987, possibly due to greasy spot, produced

a large number of short shoots during flush one. An average of 68

shoots were initiated per tree, compared to 17 and 16 in 1985 and

1986. Although these shoots in 1987 were much shorter than in the

other years, total shoot length in flush one was greater due to the

large number of shoots.

There was no uniform effect of irrigation treatment on average

shoot length or leaf size. This seems in contrast to the

well-established effect of decreased organ expansion due to water

deficits (Hsiao, 1973). However, Hanson and Hitz (1982) state that

under conditions of extreme diurnal variation in leaf water status,

expansion may be inhibited during the day, but long-term growth may

not be affected in some plants. There was little contribution of

soil water deficit to the pronounced midday decreases in xylem

potential during the early summer in this study, but a considerable

decrease in CO2 assimilation occurred with increased soil water

deficit (Chapter V). In contrast to the lack of effect on average

shoot length, treatments more consistently affected the number of


L









Table 3-7. Shoot number, average shoot length, and total shoot
length for three growth flushes of young 'Hamlin' orange trees
as related to irrigation based on soil water depletion, 1987.



Soil water Shoot Total

depletion Shoot length length

(%) no.z (cm)z (cm)



Flush one (4/12 5/6)y

20 (High) 77 4.9 376.4

45 (Mod.) 63 4.0 254.8

65 (Low) 64 4.2 264.5

Mean 68 4.4 297.4

Flush two (6/18 8/2)

20 (High) 22 16.2 362.4

45 (Mod.) 21 12.3 259.1

65 (Low) 20 11.0 216.8

Mean 21 13.1 277.4

Flush three (8/9 11/2)

20 (High) 15 32.4 486.6

45 (Mod.) 6 32.1 192.4

65 (Low) 6 27.4 153.5

Mean 9 30.6 271.7

SE 6 1.5 ns



ZMeans of 5 trees/treatment.
Range in dates of initiation.
xSE for comparison of means among growth flushes. Comparison
among irrigation levels within growth flushes is inappropriate
since irrigation x flush interaction is not significant.







57

shoots that initiated growth during the later flushes of the season.

Perhaps the level of carbon reserves at the time of flush initiation

affects the number of shoots that grow.

Average shoot length consistently increased with the three

successive flushes during all years of the irrigation scheduling

experiments, and shoot number generally decreased from flush one to

flush three. This is consistent with previous studies with citrus

growth and development in both subtropical and arid regions

(Krishnamurthi et al., 1960; Mendel, 1969).

Final Measurements

Final canopy size was not different for trees in the high and

moderate treatments in 1985 or 1986. The low treatment, however,

significantly decreased canopy volume (P<.0003), trunk cross

sectional area (P<.0001), dry weight (P<.0043), total shoot length

(P<.0040), and leaf area (P<.0029) when compared to the high

treatment in 1985 (Table 3-8). Similarly, in 1986 trees in the low

treatment had significantly less canopy volume (P<.0023), trunk

cross-sectional area (P<.0017), dry weight (P<.0086), and leaf area

(P<.0217) than in the high treatment (Table 3-8).

Irrigation treatment effects on canopy growth followed a

different pattern in 1987. Both the moderate and low treatments

reduced tree growth measured as trunk cross sectional area

(P<.0029), canopy dry weight (P<.0050), shoot length (P<.0257), and

leaf area (P<.0112) when compared to the high treatment (Table 3-8).

In addition, final canopy size for all treatments was generally

smaller in 1987 when compared to the other years. This difference

in growth response may be due to the nursery trees used in 1987. As









Table 3-8. Canopy volume, dry weight, shoot length, leaf area
and trunk cross-sectional area of young 'Hamlin' orange trees
as related to irrigation based on soil water depletion.


Trunk

cross


Soil


water

depletion

(z)


Canopy

volume

(m3)


sectional

area

(cm2)


Canopy

dry

wt.

(g)


Shoot

length

(cm)


Leaf

area

(m2)


1985


(High)

(Mod.)

(Low)



(High)

(Mod.)

(Low)



(High)

(Mod.)

(Low)


0.57

0.52
**
0.33



0.51

0.54

0.31



0.56

0.37

0.36


5.1

4.9
**
4.2



8.0

8.0
**
6.8



4.7
*
3.8

3.3
3.3


424.6

425.8
336.6
336.6


950.8

950.9
752.4
752.4


1.3

1.2
e*
1.0


1986


379.2

383.3
300.0
300.0


950.4

942.8

894.3


1.3

1.3
**
0.9


1987


393.2
**
259.0
229.6
229.6


1428.6

937.4

872.2


1.4

0.7
0
0.5


'Response is significant when compared with
depletion treatment by the Williams' method,
respectively.


the 20% soil water
5% and 1%,







59

discussed earlier, abscission of most leaves occurred shortly after

transplanting to the field, inducing a flush with more than three

times the number of shoots than occurred in 1985 or 1986. Newly-

emerged shoot growth in citrus is mainly dependent upon stored

reserves (Sinclair, 1984; Van Noort, 1969), and this heavy flush may

have severely depleted the level of reserves in these trees. Webb

(1981) stated that following a heavy defoliation of fir trees,

recovery was related to the level of carbohydrate reserves in the

trees, and trees with a low level of reserves died. Survival was

not a problem in this study, but a severe depletion of reserves

early in the spring may have caused reduced canopy growth in general

in 1987, especially in the moderate and low treatments. Weak trees

typically respond to more frequent irrigation with increased growth.

Seasonal Distribution of Shoot Growth

More than two-thirds of the trees in 1985 initiated their first

growth flush the last week in May, and all had begun growth by 25

June (Fig. 3-7). Initiation of the second flush occurred from 12

July to 19 Aug.. Irrigation treatments did not affect the dates of

initial growth in either of the first two flushes. Initiation of

the third flush occurred over a longer time period than for flushes

one and two. The entire population of trees receiving the high

irrigation treatment had begun growth of flush three in October.

The percentage of trees initiating this flush in the moderate and

low treatments was shifted to later in the season, and approximately

40% of the trees in the low treatment did not initiate a third

flush. The shapes of curves in flush three differed significantly

among treatments.


-


















100









T r ...( m*t*
* :





: ,


ns ns **
n I
50 100 150 200
Time (days)
I I I I I I I I I
J J A S O N
Time (months)








Fig. 3-7. Cumulative percentage of trees in three irrigation
treatments growing over the 1985 season. ( ) 20% soil
water depletion (SWD), ( m ) = 45% SWD, ( eeo* ) 65Z S)D.
ns,** indicates nonsignificance or significant at the 1Z level,
respectively, according to analysis of covariance test of
homogeneity of the three equations.







61

Similarly, initiation of the first growth flush in 1986 was not

affected by irrigation treatments, and occurred between 25 May and

17 June (Fig. 3-8). The two subsequent flushes were initiated over

a longer time interval, ranging from 7 July to 13 Sept. in flush two

and 30 Aug. and Dec. in flush three. The shapes of the curves in

both flushes two and three differed significantly among treatments.

Again, the percentage of trees initiating a flush in the moderate

and low treatments shifted growth to later in the season compared to

the high treatment. Approximately 25 and 35% of the trees in the

moderate and low irrigation treatments had not initiated growth of

the third flush by the end of the season.

Initiation of flush one in 1987 occurred earlier than in the

other years due to an earlier planting date (Fig. 3-9). Trees

across all treatments began growth between 12 April and 6 May.

Flushes two and three were again spread over a longer time interval

than flush one. Treatment effects were similar to those in 1986 in

that the shapes of the curves for flushes two and three differed

significantly. All trees initiated a third flush by the end of the

season, in contrast to 1985 and 1986, possibly resulting from the

earlier planting date.

Microsprinkler irrigation of these young trees at 45 and 65%

SWD clearly prolonged the period between initiation of flushes two

and three in some cases when compared to 20% SWD. Similarly, Cooper

et al. (1969) stated that a prolonging of the period of quiescence

between growth flushes of citrus commonly occurs in response to

drought. This delay in shoot growth may result from decreased

levels of available reserves, since a considerable reduction in CO2




















100





75



0O

2 50
.5



25


100
Time (days)


I I


I I


Time (months)


Fig. 3-8. Cumulative percentage of trees in three irrigation
treatments growing over the 1986 season. ( inl) 20Z soil
water depletion (SWD), ( m ) 45Z SWD, ( **.** ) 65Z SWD.
ns,** indicates nonsignificance or significant at the 1Z level,
respectively, according to analysis of covariance test of
homogeneity of the three equations.







63










Flush 1 Flush 2 Flush 3
100


I I.'
0






-75- *
0
0



*C*







so- I ."*
0


















o 100 ISO 200

Tirm (days),
A M J J A 8 0

Tinm (months)









Fig. 3-9. Cumulative percentage of trees in three Irrigation
treatments growing over the 1987 season. (- ) 20Z soil
water depletion (SWD), ( --om ) 45Z SWD, ( *e. ) 65Z SWD.
ns,*,** indicates nonsignificance or significant at the 5
and 1Z level, respectively, according to analysis of
covariance test of homogeneity of the three equations.
0












00 100 0Sa 20







64

assimilation occurred in response to increased soil water deficit

(Chapter V). Perhaps a critical level of available reserves must be

met before subsequent shoot growth begins. The period between

flushes of citrus has been shortened with longer daylengths

(Piringer et al., 1961). Both increased irrigation frequency and

daylength would fit well into a hypothesis of available reserves

controlling the length of time between flushes, as both would

increase CO2 assimilation over time. Shoot length and number of

leaves per shoot are dependent upon available reserves before growth

begins (Van Noort, 1969), and the date of growth may be dependent as

well.

Other factors may be involved with this delay in growth. The

amount of growth inhibitors reportedly decreases during the period

between flushes (Mendel, 1969). It would be of value to understand

the interactions of growth delays caused by less frequent irrigation

with the levels of growth promoters and inhibitors. Knowledge of

the interactions of root growth periodicity would also add

understanding.

Time of shoot growth initiation was uniform from plant to plant

in the spring flush all 3 years and across all irrigation

treatments, but became more widespread in the subsequent flushes.

This pattern has been previously documented with mature citrus trees

in Australia (Sauer, 1951).

Microsprinkler Spray Patterns

Distributing irrigation water over 5.9 and 9.2 m2 by using 90

and 180 spray patterns during the second growing season did not

result in different trunk cross sectional area or canopy volume







65

(Figs. 3-10, -11). Furthermore, measurements of root distribution

suggested that 90* emitters placed water over the majority of the

citrus tree's root system after one season of growth (Chapter IV).

When beginning with 900 spray patterns, there appears to be no

advantage in changing to a larger pattern after the first season in

the field. By directing more water on tree trunks, 90 patterns are

more efficient than larger patterns for freeze protection purposes

(Rieger et al., 1986). Maintaining 90 patterns for irrigation

purposes more than one season allows their use throughout another

winter for freeze protection.

In summary, growth of young 'Hamlin' orange trees was similar

in 2 out of 3 years with the high (20% SWD) and moderate (45Z SWD)

irrigation treatments, but was reduced by the low treatment (65%

SWD). The seasonal amount of water applied to the moderate

treatment averaged about 50% of the amount applied to the high

treatment. There was a pronounced delay in summer and fall growth

flushes and in some cases a reduction in the number of shoots per

tree in the low irrigation treatment.

The optimum level at which irrigations should be scheduled

cannot be precisely determined from these studies, but is most

likely between 20 and 45% SWD. These values are in general

agreement with other reports. Smajstrla et al. (1985) obtained

optimum growth of young 'Valencia' orange trees while scheduling

irrigations at 45% SWD in a field lysimeter study. Leyden (1975a)

presented no data, but suggested scheduling basin and strip watering

methods of irrigation on 30% SWD, based on field observations.

Therefore, it appears that on a per tree basis, microsprinkler






















14 3

E
10
6 E


10

2 so


900 1800 90 180
















Fig. 3-10. Trunk cross sectional area (TCA) and canopy volume
of young 'Hamlin' orange trees as influenced by microsprinkler
irrigation spray pattern (90 and 180), 1986. There were no
significant differences between patterns, 5Z level.


_ I_ _






















10 2
CO
8 E 1.6
-(D
S6 .E 1.2

< 4 0.8
0




900 1800 90 1800














Fig. 3-11. Trunk cross sectional area (TCA) and canopy volume
of young 'Hamlin' orange trees as influenced by aicrosprinkler
irrigation spray pattern (900 and 180*), 1987. There were no
significant differences between patterns, 5% level.







68

irrigation at 45% SWD is as effective as at 20% SWD, resulting in a

considerable reduction in water use. However, on a population basis

the final growth flush of some trees will be reduced at the moderate

irrigation level probably because of inherent variability in SWD

among trees within the treatment. Variability in soil moisture

content is a particular problem in flatwoods areas of Florida where

soil type is extremely variable. Consequently, growers must be

certain to monitor soil moisture content carefully in a

representative portion of their grove to ensure that the entire

irrigated area receives adequate water.















CHAPTER IV


MICROSPRINKLER IRRIGATION AND GROWTH OF YOUNG
'HAMLIN' ORANGE TREES. II. ROOT
GROWTH & DISTRIBUTION



Introduction

Perennial tree root systems respond to a variety of

environmental and management factors, including irrigation.

Irrigation method and scheduling directly affect soil water content

and indirectly affect edaphic factors like physical impedence,

fertility, and aeration, thus altering root growth and distribution.

Increases in soil water content generally increase root growth

provided that oxygen or salinity levels are not limiting.

Irrigation has been observed to increase root growth or density of

many crops (Goode et al., 1978; Goode & Hyrycz, 1970; Ponder &

Kenworthy, 1976; Richards & Cockroft, 1975), including citrus

(Bielorai, 1982; Bielorai et al., 1981; Bielorai et al., 1984;

Hilgeman & Sharp, 1970; Rodney et al., 1977). In addition to growth

responses, irrigation also affects root distribution. Frequent

irrigation commonly increases the proportion of roots in shallow

zones (Beukes, 1984; Cahoon et al.; 1961, Goode et al., 1978; Goode

& Hyrcyz, 1964; Hilgeman et al., 1969; Hilgeman & Sharp, 1970; Layne

et al., 1986; Levin et al., 1980).







70

Increasing demands on water resources require that irrigation

be used more efficiently. Micro irrigation systems are becoming

widely used in Florida, however, there is little information

available on irrigation scheduling for young citrus trees under

Florida's subtropical climate and sandy soil conditions. Therefore,

our objectives were to study the influence of three microsprinkler

irrigation treatments on root growth and distribution of young

citrus trees.

Materials and Methods

Three field experiments were conducted at the Horticultural

Research Unit near Gainesville, FL, with site and environmental

variables as described in Chapter III. Microsprinkler irrigation

was scheduled based on available soil water depletion (SWD) in the

root zone. Treatments were defined as high (20% SWD), moderate (45%

SWD), and low (65% SWD).

Root Growth and Lateral Distribution, Excavated Root Systems

Twenty, 21, and five root systems per treatment were excavated

by hand in 1985, 1986, and 1987, respectively. Excavation procedure

was described in Chapter III. Included in this group were samples

of 10 (1985, 1986) and five (1987) randomly selected plants per

treatment in which the root system was dyed prior to planting.

Dyeing was accomplished by dipping roots for 15 sec in a 1% solution

of safranin-0 as described by Kaufmann (1968). Root dry weights

were determined following oven-drying for 3 days at 80 C. Roots

which had developed during the experimental period were easily

distinguished from initial dyed roots, and were oven-dried and

weighed separately. Random samples of 10 (1985, 1986) and five









(1987) plants per treatment were utilized to determine the increase

in root volume. Water displaced by submersing root systems into

water-filled containers was collected to determine root volume at

planting time and after excavation.

Samples of 10 (1985) and five (1986, 1987) randomly selected,

excavated root systems per treatment were used to determine lateral

root distribution. After excavation these root systems were

separated into three concentric zones (0-40, 40-80, and >80 cm) from

the trunk. Distribution of total and fibrous (( 1.5 mm) roots was

determined on a dry weight and percentage basis.

Circular Trench Profiles

A modification of the trench profile method (Atkinson, 1980)

for root distribution studies was used to determine irrigation

effects on vertical root distribution at two distances from tree

trunks. The method was modified by Huguet (1973) by using a

curvilinear instead of a straight trench, based on the assumption

that roots radiate concentrically from trunks. Based on the same

assumption, circular trenches were dug by hand around each of three

randomly selected trees per treatment in Dec. 1987. The initial

trench was excavated 80 cm from the trunk to a depth of 50 cm. The

profile was smoothed with a spade, then ca. 2 cm of soil was removed

by hand with the aid of a brush to expose roots. It was not

difficult to remove a 2 cm layer of soil from the circular profile

walls due to the loose, sandy nature of the soil (Kanapaha sand).

Roots were counted at each of four depth increments (0-10, 10-20,

20-30, and >30 cm) and categorized on the basis of total root

number, and the number of fibrous (< 1.5 mm) and non-fibrous (> 1.5


. i







72

mm) roots. Vertical distribution was determined on a concentration

(roots/m2) and percentage basis.

Roots were counted to allow calculation of root length density

immediately after recording root numbers. Lengths of exposed roots

were estimated by visually counting the number of 2-cm sections of

exposed root length as described previously (Bohm, 1976) in each

depth increment. Roots that were perpendicular to the profile wall

in situ extended ca. 2 cm, and were assigned a value of one, while

roots longer than 2 cm were given values greater than one. For

example, roots which were 3, 4, and 5 cm long were given values of

1.5, 2.0, and 2.5, respectively. Summation of these values

estimated the number of 2-cm increments of exposed roots. Root

length density (mm dm-3) was determined for each depth by

calculating total root length (multiplying the sum of all values by

2 cm) and volume of the 2 cm layer of soil on each profile wall. A

second circular profile wall was subsequently exposed on the same

tree at 40 cm from the trunk and the process repeated.

Analysis of Data

Data from excavated root systems were analyzed separately for

the 3 years. Data on root dry weights were subjected to analysis of

variance and data on root volumes to analysis of covariance to

standardize initial root volumes. Where measurements were

significantly different among irrigation treatments, Williams'

(Williams, 1971) test was used to compare means.

Lateral distribution of total and fibrous (< 1.5 mm) root dry

weights were analyzed separately by a split-plot analysis with

irrigation treatments as main plots and lateral zones as subplots.







73

Root percentage distribution among zones was analyzed in the same

manner following arc sine transformation.

Measurements from profiles in 1987 of root density and root

concentration of fibrous, non-fibrous, and total roots were analyzed

separately. Percentage distribution data were transformed using an

arc sine transformation prior to analysis. All data sets were

analyzed by a split-split plot analysis with irrigation treatments

as main plots, distance from the trunk as sub-plots, and depth

increment as sub-sub-plots.

Results and Discussion

Root Weight and Volume Measurements

Irrigation treatments had variable effects on root dry weight

and volume from year to year (Table 4-1). Dry weights were not

affected in 1985, but root volume was decreased significantly at the

low compared to the high irrigation treatment (P<.0083). Irrigation

levels significantly affected root growth in 1986 and 1987. Trees

receiving the low irrigation treatment in 1986 had less total root

dry weight (P<.0052), new root dry weight (P<.0249), and root volume

(P<.0116) than those in the high treatment. Both moderate and low

treatments in 1987 decreased total root dry weight (P<.0170), new

root dry weight (P<.0455), and root volume (P<.0345) compared to the

high treatment.

Root volume ranged from 850 to 1190 cm3 in 1985 and 1986, and

from 550 to 860 cm3 in 1987 (Table 4-1). Root dry weight ranged

from 260 to 350 g in 1985 and 1986, and from 160 to 260 g in 1987.

These values were in close agreement with those of Bevington (1983)

and Bevington and Castle (1982) for 13-month-old 'Valencia' orange













Table 4-1. Total root dry weight, dry weight of new roots,
and root volume of young 'Hamlin' orange/sour orange
trees as related to irrigation based on soil water
depletion.


Soil water Total root New root Root
depletion dry wt dry wt volume
(%) (g) (g) (cm3)


1985 (n=20)
20 (High) 348.7 156.3 1189.2
45 (Mod.) 333.9 158.8 1058.4
**
65 (Low) 297.3 139.4 920.5
1986 (n-21)
20 (High) 342.1 191.0 1180.9
45 (Mod.) 337.8 185.4 1396.5
**
65 (Low) 262.1 128.2 854.3
1987 (n-5)
20 (High) 257.8 126.2 858.2
45 (Mod.) 169.4 84.8 559.7
65 (Low) 159.6 81.2 554.0


Response is significant when compared with the 20%
SWD treatment by the Williams' method; 5% and 1%
levels, respectively.







75

on 'Carrizo' citrange or rough lemon trees grown in root observation

chambers. Tree age in this study was about 7 months in 1985 and

1986, and 8 months from planting in 1987. The comparable root sizes

in the field and root chamber studies despite the much shorter

growing period in the field study may be due to rootstock

differences or to the allowance of unrestricted root development

under the field conditions.

Shoot:root ratio was calculated using dry weight measurements

reported in Chapter III (Table 3-8). The SWD ranges used in this

study did not significantly influence shoot:root ratio, which ranged

from 1.16 to 1.54, with a mean of 1.30. These values are lower than

those reported for 1 1/2-year-old mandarin seedlings in India, where

healthy trees had a shoot:root ratio of 1.92 (Aiyappa & Srivastava,

1965). Comparatively, excavated mature citrus trees in California

had a ratio of about 3.5 (Cameron, 1939; Cameron & Compton, 1945),

and in Florida a ratio of about 2.2 (Castle, 1978).

Root growth was less in 1987 compared to the other years

despite a longer growing season. Furthermore, trees under moderate

irrigation grew similarly to those in the high treatment in 1985 and

1986, but had less root growth than those in the high treatment in

1987. These discrepancies may be due to the nursery trees used in

1987. Excessive leaf abscission occurred shortly after

transplanting trees to the field in 1987, inducing a growth flush

with a large number of shoots per tree (Chapter III, Table 3-7).

Such a growth flush early in the spring may have depleted available

reserves thus decreasing subsequent root growth.









In all 3 years root growth was decreased at the lowest soil

moisture content. Similarly, decreased root growth in response to

decreased soil moisture has been frequently reported for citrus

trees of various ages (Bevington & Castle, 1985; Bieloral, 1982;

Bielorai et al., 1981; Hilgeman & Sharp, 1970; Rodney et al., 1977).

Lateral Root Distribution

Lateral distribution of roots on a dry weight basis was

affected by irrigation treatments. The irrigation treatment x

lateral zone interaction for total root weight was significant in

1985 (P<.0238), 1986 (P<.0405), and 1987 (P<.0141). A greater

quantity of roots within 40 cm of the trunk occurred in response to

the high treatment when compared to moderate and low treatments

(Tables 4-2, -3, -4). Root weight in the 40-80 and >80 cm zones was

not affected by irrigation, but in all 3 years the high treatment

produced more roots than other treatments in the 0-40 cm zone. In

contrast, irrigation treatments did not affect lateral distribution

of fibrous (<1.5 mm) root dry weight or distribution of both fibrous

and total roots on a percentage basis.

Across all irrigation treatments and years root dry weights and

percentages decreased significantly (P<.0001) with increased

distance from trunk (Tables 4-2, -3, -4). One exception occurred in

1986, when fibrous root weight was not different among the lateral

tones (Table 4-3). The percentage of fibrous roots within 40 cm of

the trunk ranged from 40 to 60%, between 40 and 80 cm from 26 to

39%, and >80 cm from 13 to 29%. A much larger percentage of total

root weight was located close to the tree, with 68 to 84%, 11 to


_i


f
























N








F.'
0


0


0
M
W



I-






F.
H




s^

x
i^


*




0 b
41



1n
* -*l







- 0
vIIka
0
0 0



dk W


0O









wo




0 w

05 0%
It 41


o 0



40

b,4 0 0


S-S 0




5 .0



u40

*0 10
>0B




I-'


v


%-I
60


0




Q


%O'.0%O
* *


cvQ Cl



* *I
%Or-VLr


co s0 00 C
* *
C C4M -' t
CNrCO


* *
rCZ NA
Ln P-


*(V M
N N Ob
CNrl


* *
Cn

%0 1 0 I
N *-aCO


* *
"*o ine


-aNNh
0000
N-r%~


* *
000




* *
,4 ,,U ,




O -,ON


, ,-0 0
* *
-1-Cch4
(V.'l


000 -

0* **


I 0
I 4J



r54
0


0


I
4.1
0


* *
o if. N
4<< --
M


00%
* *
00-'
Oc so
aCD M


* *


I *4
I .ri '
0o 4i


000
* *

00


00 -. 0 0 V-4
TOO Go 0 VO go 0 >oo%
I I ca41 I O414
SOOA O 0 W
OO 1OT 1o 4


* c

0 -r4 0
N bo N



0a w0

0 cOc

1P4





S0
4 k
* *..I
000


r4

0 0

410C



00


0 CO
W 0
0 0 al
- H *




w W4


0 0 0O4
o u 4



00 0
000
N W




0 0 0
). 5B B


4'440.0
U D4
&r


F.
H


H
U, -
bO


___



























J-w


Vik

* cc
0 a
kO
v4

44 4)
00

.0 00


1A 0

.4 w


0 bp





- 4 .
0 o




O OD
S0 00



3 h -

0 0





1 0 $
r4,O
.O
' 0


.0





I 0 0

0
IU -


u g$

10 44
.14
9z


* 9
i^ r


N*

'T f


* 4 T L M
A** C4
M C4 CN co
hON N cO


N
U)








0



V

v










b0

I H












is
v6










3
<-
60
^


0-to
* *
r r sO
CMCM


C %O
* *
0 o o
-tcCMJ


Nf so *
* 9 S 5 |

0N N 00%





* I



0.4







c ^o
00 p.4
I 0O w
0 0 A 0
-t M I-.
O~O H


000
000
SCN)
0 a 0


\04 0C4
0000




s0) 0







* *
N S0

0> ^CT


00'.o
* 9 9
%0UC4


0I 0
P t^-


C00.i














NC44
* *


1(4
0
N










O
* *













* 0
O m
^ ^


S00 o
O ODA C00
1I 1 -
CoA o \oi


* *
00
*av <


* -



bO
0 K

0 0


4 ObO
0 4o
60



0 0
0 0b



S*00



40 60


C u




ew a4
40 c
oa a
N P 0a






000

0 0 a O
660
0 00




0





111bo
M a e
e N

































0 0







0 000





4 0
vw
to

















l0 0 0
to 44
C 60
0 0.





















00k'
> 4
U 0
N3 0




























$44
u '.















1a 0 0
(0
w0 H
O 0


4 0. a
















-k
E -4






a 4a











.5.4
-ao
pa00
.04


0O 00 00
*** 4


m *
oo) --T
*

10 C00 0 0
C14e.C in


N













bo






c





43





0








41 bo

ka
V4M
k v





0
W










M








t* 4 4
<0 m
s


)
r)


400
* *
0% I
- -l


%0 0- -
'.orM u


0-4.4
'.cccD


00 -D O
I I 0c W
0 0 A 0 CO
? E-4


0 0 0
* *
n 0 in m



---



r- r-



* 4
un c -4




-4

* S S
a0'.0u,


00
0 *
C 000
C..'


-1

cr


u

\O
*
'n
S
'c
0%
-4





I
I




< n ^
**

-I


~I


I c- 0 .
I.-


-4


I '.r-


00
* .
P-.4
0%4


* I
'O I




* *
co Go Ln

NM
'.,'cv


00 -4
T 00 0 %
0 0 A 0
4 E-4


0 < N








A 00 C
C-4





0; C
owe c
*** ,


.0 O

N 00 N





0 rf


C:
4 C0
Sdk









4 0. ar.

uH N
e 4






a 05
0 a
00

44 4- O
430.







kN 00
B0 0
OaE






N 4






00B C







80

212, and 4 to 11% of root dry weight in the three concentric lateral

zones, respectively.

The decrease in root growth with increasing distance from the

trunk has been reported for citrus trees of all ages (Aiyappa &

Srivastava, 1965; Bielorai, 1977, 1982; Bielorai et al., 1981;

Cahoon et al., 1964; Minessy et al., 1971; Yagev & Choresh, 1974).

Healthy 1 1/2-year-old mandarin trees in India had 60.0, 14.7, 13.2,

and 12.1% of their root dry weight in 0-30, 30-60, 60-90, and >90 cm

concentric zones, respectively (Aiyappa & Srivastava, 1965). These

values are similar to those from the young 'Hamlin' orange trees in

this study.

Maximum distance of lateral root spread pooled over the 3

seasons averaged 137 cm for the high treatment, compared with 127

for moderate and 121 cm for the low treatments, respectively (data

not shown). In comparison, 2-year-old 'Valencia' orange trees in

Australia had a lateral root spread of about 150 cm (Till & Cox,

1965), and 1 1/2-year-old mandarin trees in India had a spread of

160 cm (Aiyappa & Srivastava, 1965). Savage et al. (1945) observed

that 6-year-old trees budded on common rootstocks and spaced 30 x 90

cm in a Florida nursery had lateral root spreads up to 150 cm.

There is considerable discussion about the optimum irrigation

pattern to use for young citrus trees in Florida. Roots covered an

area of about 5.2 m2, with 64% of this area covered by the 90*

emitters in this study. Furthermore, 96% of the roots on a dry

weight basis were within the wetted zones due to the high

concentration of roots within 80 cm of the trunk (Tables 4-2, -3,

-4). These data imply that 90 microsprinkler emitters cover the







81

majority of the young tree's roots after one season of growth. Tree

growth comparisons utilizing 90 and 180 emitters during the second

season of growth also suggest 90 emitters are adequate for up to 2

years in the field (Figs. 3-10, -11).

Weight of fibrous roots accounted for 27% of the total. This

value is similar to those found on root systems of rough lemon and

'Carrizo' citrange in the nursery (Bevington & Castle, 1982), but

considerably less than found on healthy 1 1/2-year-old mandarin

seedlings (37.5%) growing in India (Aiyappa & Srivastava, 1965). In

comparison, feeder roots on 8- to 9-year-old orange trees on sweet

orange rootstock accounted for less than 15% of the root system dry

weight (Cameron & Compton, 1945).

Circular Trench Profiles

Analysis of variance results showing effects of irrigation

treatments, distance from tree trunks, and depth increments on root

density (mm dm3) and concentration (roots m-2) of total, fibrous,

and non-fibrous roots are summarized in Table 4-5. A larger

proportional increase in total root concentration occurred at 40

than at 80 cm from the trunk in response to the high treatment

(Table 4-6). This corroborates results of irrigation treatment

effects on lateral root dry weight distribution (Tables 4-2, -3,

-4). The high irrigation treatment produced a proportionally larger

increase in root number in the 10-20 cm layer than in the other

layers (Table 4-6). The greatest number of roots grew in this

layer, regardless of irrigation treatment and there was little root

growth in the 0-10 cm depth regardless of treatment, and trees under

the high treatment had the highest root concentration at all depths


rSI













Table 4-5. Partitioning of variance into main and interaction
effects for 1987 circular trench profile root growth variables of
'Hamlin' orange trees as described in Tables 4-6, 4-7, and 4-8.



F tests on variance (MS) ratios

Total Fibrous Non-fibrous Total

root root root root

Source df conc. cone. cone. density
s *
Irrigation (I) 2 5.39 6.05 1.10 5.73
*** *** *** ***
Distance (Di) 1 55.40 54.88 44.03 55.60

I x Di 2 5.90 6.40 1.55" 6.10

Depth (De) 3 60.12 60.08 38.06 58.58

I x De 6 5.76* 6.11** 1.71n 5.32*

Di x De 3 21.61 21.35 15.79 19.91

Ix Di x De 6 2.27ns 2.43ns 0.92us 2.21ns


ns,*,**,*** Nonsignificant or significant F values at the 52,
1%, and .1% levels, respectively.













Table 4-6. Total root concentration at four depths and two distances
from trunk of young 'Hamlin' orange/sour orange trees as influenced
by irrigation based on soil water depletion (SWD) as determined
using circular trench profiles, 1987.



Irrigation treatment

Depth 20% SWD 45% SWD 65% SWD

(cm) High Mod. Low Mean
-2 z
Root cone. (roots m ) at 40 cm

0-10 19.9 37.2 11.9 23.0

10-20 965.5 655.2 299.7 640.1

20-30 447.0 244.0 202.9 298.0

>30 291.8 156.5 70.3 172.9

Mean 431.1 273.2 146.2 283.5y
-2 Z
Root cone. (roots m ) at 80 cmz

0-10 13.9 6.6 0 6.8

10-20 215.5 222.8 45.1 161.1

20-30 159.1 172.4 87.5 139.7

>30 65.7 24.6 43.1 44.5

Mean 113.6 106.6 43.9 88.0y


Zn-3 for irrigation x distance x depth means.
YMean for distance from trunk.









below 10 cm. This increase in root concentration in response to

maintaining higher soil moisture levels corroborates the dry weight

measurements made on excavated root systems (Table 4-1).

The concentration of fibrous roots (<1.5 mm) followed patterns

similar to total root concentration in regard to irrigation

treatments, distance from trunk, and depth because fibrous roots

comprised the majority of the total roots (Table 4-7). Differences

in non-fibrous (>1.5 mm) root concentration were not associated with

irrigation treatments. There were four times as many large roots at

40 cm from tree trunks than 80 cm, and almost half were located

between 10 and 20 cm in depth (Table 4-8). Trees receiving the low

irrigation treatment had 10 times as many fibrous as large roots;

whereas, trees in the high and moderate treatments developed

approximately 15 times the number of fibrous as large roots.

The percentages presented in Table 4-9 were pooled over the two

lateral distances since they did not interact with percentage

distribution with depth. Greater than 50% of the root concentration

was between 0-20 cm with high and moderate irrigation, but was

deeper than 20 cm with low irrigation. Because of the small number

of replications and the variability, this difference was not enough

to be significant. Across all treatments, percentages of roots were

significantly (P<.0001) different with depth (Table 4-9). Roughly

half of the roots were 10-20 cm deep, with 35% between 20-30 cm, and

less than 15Z below 30 cm. More than 85% of the roots were located

between 0-30 cm in depth.

The concentration of roots in the top 30 cm after 1 season of

growth suggests that when soil moisture monitoring is used for













Table 4-7. Concentration of fibrous (<1.5 mm) roots at four depths
and two distances from trunk of young 'Hamlin' orange/sour orange
trees as influenced by irrigation based on soil water depletion
(SWD) as determined using circular trench profiles, 1987.



Irrigation treatment

Depth 20% SWD 45% SWD 65% SWD

(cm) High Mod. Low Mean

Root cone. (roots m-2) at 40 cmz

0-10 19.9 33.2 11.9 21.7

10-20 903.2 614.1 263.9 593.7

20-30 415.1 228.1 181.7 275.0

>30 279.9 149.9 65.0 164.9

Mean 404.5 256.3 130.6 263.8y

Root cone. (roots 2) at 80 cm

0-10 13.3 6.0 0 6.4

10-20 199.6 211.5 43.1 151.4

20-30 150.5 157.8 84.9 131.1

>30 63.0 23.9 41.8 42.9

Mean 106.6 99.8 42.5 83.07


Zn-3 for irrigation x distance x depth means.
YMean for distance from trunk.













Table 4-8. Concentration of non-fibrous (>1.5 mm) roots at four
depths and two distances from trunk of young 'Hamlin' orange/sour
orange trees as influenced by irrigation based on soil water
depletion (SWD) as determined using circular trench profiles, 1987.



Irrigation treatment

Depth 20% SWD 45% SWD 65% SWD

(cm) High Mod. Low Mean

Root cone. (roots m-2) at 40 cm

0-10 0 4.0 0 1.3

10-20 62.3 41.1 35.8 46.4

20-30 31.8 15.9 21.2 23.0

>30 11.9 6.7 5.3 8.0

Mean 26.5 16.9 15.6 19.7y
-2 z
Root cone. (roots -2) at 80 cm

0-10 0.7 0.7 0 0.5

10-20 15.9 11.3 2.0 9.7

20-30 8.6 14.6 2.7 8.6

>30 2.7 0.7 1.3 1.6

Mean 7.0 6.8 1.5 5.1y



Zn=3 for irrigation x distance x depth means.
7Mean for distance from trunk.


; '













Table 4-9. Percentage distribution of root concentration with depth
of young 'Hamlin' orange/sour orange trees as influenced by
irrigation based on soil water depletion (SWD). Percentages are
means of measurements from 40 and 80 cm distances from trunk, 1987.



Irrigation treatment

Depth 20% SWD 45% SWD 65% SWD

(cm) High Mod. Low Mean

<1.5 mm

0-10 1.4 2.6 1.6 1.8

10-20 47.9 57.2 38.0 47.7

20-30 34.1 30.6 41.4 35.4

>30 16.6 9.6 19.0 15.1

>1.5 mm

0-10 0.6 3.1 0 1.2

10-20 50.8 48.4 45.2 48.2

20-30 36.3 38.9 38.5 37.9

>30 12.3 9.6 16.3 12.7

Total

0-10 1.3 2.7 1.4 1.8

10-20 48.1 56.5 38.4 47.7

20-30 34.2 31.3 41.6 35.7

>30 16.4 9.5 18.6 14.8







88

irrigation scheduling, measurements should be concentrated in this

zone. Furthermore, irrigation times should be limited to replenish

soil moisture only in these shallow zones. Irrigations of short

duration are adequate when using systems which direct applications

to a small areas on sandy soils. Soil moisture was monitored at a

depth of 30 cm in this study, and at 20% SWD the 90, 38 liter/hr

emitters required approximately 1 hr to replenish the root zone to

field capacity.
-3
Root length density (mm dm-3) at two distances from tree trunks

and four soil depths as influenced by irrigation treatment followed

patterns similar to root concentrations. The greatest difference

between treatments occurred between 10-20 cm, with differences of

less magnitude deeper in the profile (Table 4-10). Densities ranged
-3
from 0.5 mm dm3 at 80 cm from the trunk on trees under the low

irrigation treatment to 9.8 mm dm-3 at 40 cm from the trunk on trees

under the high irrigation treatment.

Mature citrus responded to frequent irrigation with an

increased proportion of roots in shallow root zones and a decrease

in root growth in deep root zones using flood (Cahoon et al., 1961;

Cahoon et al., 1964; Hilgeman & Sharp, 1970) and drip (Ruggiero &

Andiloro, 1984) irrigation systems. Results from this study

indicated that irrigating more often in the high treatment had no

effect on root growth in the 0-10 cm depth, and increased the

proportion of root growth in the 10-20 cm depth compared to deeper

zones. However, in contrast to the previous studies, there was no

decrease in root concentration or length density in the deeper zones

with the frequent irrigation. These results are similar to those













Table 4-10. Root length density at four depths and two distances
from trunk of young 'Hamlin' orange/sour orange trees as related
to irrigation based on soil water depletion (SWD) as determined
using circular trench profiles, 1987.



Irrigation treatment

Depth 20% SWD 45% SWD 65% SWD

(cm) High Mod. Low Mean

Density (mm dm-3) at 40 cma

0-10 0.2 0.4 0.1 0.2

10-20 9.8 6.7 3.1 6.5

20-30 4.9 2.7 2.2 3.3

>30 3.5 1.9 0.9 2.1

Mean 4.6 2.9 1.6 3.0y

Density (mm dm ) at 80 cmz

0-10 0.1 0.1 0 0.1

10-20 2.2 2.3 0.6 1.7

20-30 1.8 1.9 1.0 1.5

>30 0.8 0.3 0.5 0.5

Mean 1.2 1.2 0.5 1.



Zn=3 for irrigation x distance x depth means.
YMean for distance from trunk.


~