EPA-600/2-77-134
July 1977
Environmental Protection Technology Series
MINIMIZING SALT IN RETURN FLO1
THROUGH IRRIGATION MANAGEMEN
Robert S, Kerr Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Ada, Oklahoma 74820
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1 Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-77-134
July 1977
MINIMIZING SALT IN KETURN FLOW THROUGH IRRIGATION MANAGEMENT
Interim Report
by
U.S. Salinity Laboratory Staff
Agricultural Research Service
U.S. Department of Agriculture
Riverside, California 92502
Interagency Project No. EPA-IAG-D4-0370
Project Officer
James P. Law, Jr.
Source Management Branch
Robert S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OKLAHOMA 7.4820
-------
DISCLAIMER
This report has been reviewed by the Robert S. Kerr Environmental
Research Laboratory, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection Agency,
nor does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
ii
-------
FOREWORD
The Environmental Protection Agency was established to coordinate
administration of the major Federal programs designed to protect the
quality of our environment.
An important part of the Agency's effort involves the search for
information about environmental problems, management techniques and new
technologies through which optimum use of the Nation's land and water
resources can be assured and the threat pollution poses to the welfare
of the American people can be minimized.
EPA's Office of Research and Development conducts this search through
a nationwide network of research facilities.
As one of these facilities, the Robert S. Kerr Environmental Research
Laboratory is responsible for the management of programs to: (a) investigate
the nature, transport, fate and management of pollutants in groundwater;
(b) develop and demonstrate methods for treating wastewaters with soil and
other natural systems; (c) develop and demonstrate pollution control tech-
nologies for irrigation return flows; (d) develop and demonstrate pollution
control technologies for animal production wastes; (e) develop and demon-
strate technologies to prevent, control or abate pollution from the petroleum
refining and petrochemical industries; and (f) develop and demonstrate
technologies to manage pollution resulting from combinations of industrial
wastewaters or industrial/municipal wastewaters.
This report contributes to the knowledge essential if the EPA is to
meet the requirements of environmental laws that it establish and enforce
pollution control standards which are reasonable, cost effective and
provide adequate protection for the American public.
Uta^^
William C. Galegar '
Director \
Robert S. Kerr Environmental
Research Laboratory
111
-------
PREFACE
Maintaining or improving water quality is a prime objective for the
United States and an issue of considerable importance worldwide. In arid
areas where food production depends on irrigated agriculture, the relation
between irrigation and water quality is often dominant.
The Colorado River exemplifies the complexity of problems encountered
in a highly developed major river basin. Technological solutions are sought
for problems created by increasing salinity as use of the River's water
continues to intensify problems of economics, resource conservation, and
international comity across a wide spectrum of user interests.
The U.S. Salinity Laboratory is concerned with all aspects of salinity
as it affects agricultural production and with the effects of agricultural
operations on water quality.
This report deals with a study to evaluate the potential of modifying
irrigation management to obtain higher irrigation water use efficiency and
thus reduce the adverse effects of irrigation on water quality degradation.
Application of this concept must always be carefully related to the specific
situation. At the site chosen for these studies, the Wellton-Mohawk Divi-
sion of the Gila Project in Arizona, any reduction in return flow achieved
through irrigation management would result in an equal reduction in drain-
age water volume to be desalted by a planned desalination complex.
Jan van Schilfgaarde
Director
U.S. Salinity Laboratory
IV
-------
ABSTRACT
Two field experiments are being conducted in southwestern Arizona to
investigate the potential of reducing the salt load in irrigation return
flow by decreased leaching. Three leaching treatments of 5, 10, and 20%,
replicated nine times for citrus and five times for alfalfa, were estab-
lished and compared with conventional flood irrigation management.
Three years' results on citrus indicate that leaching percentages of
8, 11, and 22 were achieved, compared to 47% on the border flood check.
The best estimate of the annual evapotranspiration of citrus is 1400 mm.
To date, reduced leaching has not adversely affected fruit quantity or
quality. If leaching were reduced to 20%, the volume of drainage from the
3000 ha of citrus in the district would be decreased 43.7 x 106 m3/yr and
the salt load would be cut by 45,500 Mg/yr.
Leaching percentages of about 3, 5, and 10 have been obtained in the
alfalfa after 19 months. The level-basin flood check received the same
amount of water as the high leaching treatment. Results indicate that the
sprinkled plots were underirrigated and that the annual evapotranspiration
for alfalfa is about 2000 mm. Yields from the sprinkled plots have been
16% less than those on the flooded field because of underirrigation, weed
problems, and soil compaction. Even with reseeding and less frequent irri-
gation, it is unlikely that substantial improvement is possible over the
low leaching obtained on the level-basin flood check.
This interim report was submitted in fulfillment of Interagency Agree-
ment No. EPA-IAG-D4-0370 by the U.S. Salinity Laboratory, USDA-ARS, under
the sponsorship of the U.S. Environmental Protection Agency. This report
covers the period from December 5, 1973 through December 4, 1976. The
agreement has been extended through December 1978, at which time the final
results will be reported.
V
-------
CONTENTS
Foreword iii
Preface iv
Abstract v
Figures viii
Tables xi
Selected abbreviations and conversions xiv
Acknowledgments xv
1. Introduction 1
2. Summary and Conclusions 4
3. Citrus 6
Results 19
Discussion 50
4. Alfalfa 53
Results 59
Discussion 84
References 87
Appendices
A. Additional data for citrus 89
B. Additional data for alfalfa 105
vii
-------
FIGURES
No. Page
1 Location of citrus and alfalfa minimum leaching experiments in 3
southwestern Arizona.
2 Design of minimum leaching experiment on mature Valencia orange 7
trees in southwestern Arizona.
3 Illustration of instrumentation under center tree of three of the 10
nine replications for each leaching treatment. The remaining
replications have two sets of salinity sensors and one set of
tensiometers.
4 Sketch of dual-chamber, bubbler, and drip irrigation systems. 11
5 Schematic of vacuum extractor installation. 13
6 Sampling pattern for soil chloride determinations beneath the 15
center tree of plots L7 and H4 in 1975 and 1976.
7 Computed values of electrical conductivity with soil depth for 16
Valencia orange trees irrigated with Colorado River water at 5,
10, and 20% leaching.
8 Relationship between soil matric potential and hydraulic conduc- 20
tivity for Dateland fine sandy loam soil.
9 Average daily water application for the three leaching treatments 21
and the flood check by month from February 1974 to September 1976.
10 Comparison of the daily rate of pan evaporation and evapotranspi- 24
ration estimated from the amount of water applied for Valencia
orange trees in southwestern Arizona. Data averaged by month
from January 1974 to September 1976.
11 Salinity trends with time for the three leaching treatments of the 27
citrus experiment at various soil depths.
12 Time-averaged soil salinity distributions with soil depth for the 28
initial, two intermediate time periods, and the projected final
conditions.
viii
-------
No. Page
13 Average soil matric potential profiles for the three leaching 29
treatments in the citrus during 1974 and 1975.
14 Composite cross sections of soil chloride distribution under the 31
center tree of a 5% (L7) and 20% (H4) leaching plot after 1 and
2 years.
15 Frequency distribution of the ratio of the difference between in- 34
dividual chloride measurements and the mean to standard deviation.
16 Chloride-derived and calculated travel times for a parcel of water 37
to pass through the soil profile as a function of leaching frac-
tion and rooting depth.
17 Design of minimum leaching experiment for alfalfa in southwestern 54
Arizona.
18 Spray system for one section of the lateral-move irrigation system 56
serving one replication.
19 Location of instruments for two of the six plots instrumented 57
in the alfalfa experiment.
20 Relationship between soil matric potential and hydraulic conduc- 61
tivity for Indio fine sandy loam.
21 Total water application to the three leaching treatments and the 62
flood check for the alfalfa experiment.
22 Comparison of the daily rates of pan evaporation and evapotranspi- 64
ration estimated from the amount of water applied to alfalfa in
southwestern Arizona. Average of data from January 1975 to July
1976.
23 Salinity trends with time for the 10% leaching treatment of the 66
alfalfa experiment at various soil depths.
24 Time-averaged salinity distributions with depth for three leach- 67
ing treatments and the flood-irrigation check of the alfalfa
experiment.
25 Hydraulic potential distribution for the three leaching treatments 68
in the alfalfa at selected times during 1975.
26 Hydraulic potential distributions for the three leaching treatments 70
in the alfalfa at selected times during 1976.
27 Accumulated leaching percentages based on leachate from vacuum 73
extractors and suction lysimeters for the three alfalfa leaching
treatments.
ix
-------
No. Page
A-l Distribution of total head under the center citrus tree of plot 94
H4 on January 22, 1976.
A-2 Distribution of total head under the center citrus tree of plot 95
H4 on March 23, 1976.
A-3 Distribution of total head under the center citrus tree of plot 96
H4 on April 28, 1976.
A-4 Distribution of total head under the center citrus tree of plot 97
H4 on August 24, 1976.
A-5 Distribution of total head under the center citrus tree of plot 98
H4 in the morning of August 28, 1976.
A-6 Distribution of total head under the center citrus tree of plot 99
H4 in the evening on August 29, 1976.
A-7 Distribution of total head under the center citrus tree of plot 100
H4 in the afternoon on September 1, 1976.
A-8 Distribution of total head under the center citrus tree of plot 101
H4 on September 4, 1976.
A-9 Distribution of total head under the center citrus tree of plot 102
H4 on September 9, 1976.
A-10 Cross section of soil chloride distribution under the center tree 103
of a 5% (L7) leaching plot after 1 and 2 years.
A-ll Cross section of soil chloride distribution under the center tree 104
of a 20% (H4) leaching plot after 1 and 2 years.
B-l Salinity trends with time for the 5% leaching treatment of the 106
alfalfa experiment.
B-2 Salinity trends with time for the 20% leaching treatment of the 107
alfalfa experiment.
B-3 Accumulated values for irrigation, drainage, and leaching fraction 108
for the 5% alfalfa leaching treatment.
B-4 Accumulated values for irrigation, drainage, and leaching fraction 109
for the 10% alfalfa leaching treatment.
B-5 Accumulated values for irrigation, drainage, and leaching fraction 110
for the 20% alfalfa leaching treatment.
-------
TABLES
No. Page
1 Alterations of the projected electrical conductivity of the soil 17
water (S/m) at the 0.3-m depth to achieve the desired leaching
in the citrus experiment.
2 Representative soil properties of Dateland fine sandy loam soil. 22
3 Annual irrigation, rainfall, and pan evaporation for Valencia 23
orange trees.
4 Comparison of various estimates of evapotranspiration of Valencia 25
orange trees in southwestern Arizona.
5 Leaching percentages for two plots of the citrus experiment as 33
calculated from the chloride concentrations in the soil and the
irrigation water.
6 Mean and standard deviations (S) of in situ chloride concentra- 33
tions for original and &n transformed data.
7 Chloride concentration and water loss distribution under center 35
trees of 5 and 20% leaching treatment plots.
8 Cumulative drainage and leaching percentage since January 1974 39
and EC of leachates for the vacuum extractors in the three leach-
ing treatments in the citrus experiment.
9 Average drainage water compositions from the vacuum extractors 41
for the citrus experiment in June 1976.
10 Composition of drainage waters from selected vacuum extractors 42
in the citrus experiment as a function of time.
11 Predicted drainage water compositions for citrus treatments. 43
12 Average annual Valencia orange yield (kg/tree) for the minimum 44
leaching treatments and the border- and check-plot trees.
13 Average number of Valencia orange fruit harvested per tree for 44
the minimum leaching treatments and the border- and check-plot
trees.
14 Valencia orange fruit quality for 1974, 1975, and 1976 harvests. 45
xi
-------
No. Page
15 Valencia orange juice quality for 1974, 1975, and 1976 harvests. 47
16 Mineral composition of Valencia orange leaves sampled September 48
1975.
17 Average annual increase in the trunk circumference (mm) of 49
Valencia orange trees for the minimum leaching experiment in
southwestern Arizona.
18 Comparison of the percent soil oxygen at the 0.45-m soil depth 49
and 1.5 m out from the Valencia orange tree trunk for several
irrigation treatments.
19 Average percent carbon dioxide in soil air under mature citrus 50
trees as a function of time and soil depth.
20 Estimate of citrus evapotranspiration (ET) based on chloride 51
concentration from vacuum extractors or soil sampling.
21 Representative soil properties for the Snyder Ranch alfalfa field. 60
22 Comparison of various techniques of estimating evapotranspiration 63
of alfalfa in southwestern Arizona.
23 Drainage for each vacuum extractor and suction lysimeter in the 72
alfalfa experiment after January 1, 1975 until July 26, 1976.
24 Compositions of drainage waters for alfalfa field, April 1976. 75
25 Changes in compositions of drainage waters of alfalfa field 76
with time.
26 Predicted drainage water compositions for the alfalfa treatments. 78
27 Alfalfa yield (Mg/ha) for each replication of the leaching treat- 79
ments and for three locations within the flood check for the
1975 season.
28 Alfalfa yield (Mg/ha) for each replication of the leaching treat- 80
ments and for three locations within the flood check for the
first five cuttings of 1976.
29 Mineral composition of alfalfa. 81
30 Depth to water table in alfalfa experiment (m). 82
31 Percent soil oxygen at the 0.45-m soil depth in the alfalfa 83
experiment.
xii
-------
No. Page
32 Average percent carbon dioxide in soil air under alfalfa as a 84
function of soil depth.
A-l Description of soil at the citrus experimental site provided 89
by the Soil Conservation Service.
A-2 Average annual Valencia orange yield (kg/tree) by replication for 91
the minimum leaching experiment and the border- and check-plot
trees in southwestern Arizona.
A-3 Valencia orange leaf analysis summary for September 1974. 92
A-4 Annual measurement of Valencia orange tree trunk circumference 93
(mm) averaged by replication for the minimum leaching experiment
in southwestern Arizona.
B-l Description of soil at the alfalfa experimental site provided by 105
the Soil Conservation Service.
xiii
-------
SELECTED ABBREVIATIONS AND CONVERSIONS
Symbol
m
ha
£
kg
Mg
s
kPa
kW
mmole
i/s
mg/£
S/m
meq/ H
Cl
Cl
dw
iw
Cl
EC
e
ET
LF
S
SAR
TA
TDS
Meaning
meter
hectare
liter
kilogram
megagram
second
kilopascal
kilowatt
millimole
liters per second
milligrams per liter
Siemens per meter
milliequivalents per liter
chloride concentration in meq/£
chloride concentration of drainage water
chloride concentration of irrigation water
electrical conductivity of saturation extract
evapotranspiration
leaching fraction
standard deviation
sodium-adsorption ratio
total acid
total dissolved solids
Equivalent
3.3 feet
2.47 acres
0.26 gallons
2.205 Ibs
tonne, 2205 Ibs
0.01 bar, 0.15 Ib per square inch
1.34 horsepower
10 3 gram-formula weight
15.8 gallons per minute
1 part per million
10 millimhos per centimeter
xiv
-------
ACKNOWLEDGMENTS
The work reported here could not have been carried out without the
ready cooperation of a number of individuals and organizations.
Mr. C. V. Spencer, President, Spencer and Spencer, Inc., managers of
the Desert Valencia Ranch, made available the orchard where the citrus ex-
periment is conducted and has assisted in numerous ways. His support and
expertise have been invaluable to the success of the work and his continu-
ing enthusiasm is infectious.
Similarly, Mr. W. M. Wootton, Manager of the Snyder Ranch, provided
the land for the alfalfa experiment, and arranged for land preparation,
seeding, spraying, and harvesting. Without his continuing advice and sup-
port, the work could not have been done.
The Wellton-Mohawk Irrigation and Drainage District has cooperated
fully in making special arrangements for water supply and power. Its
cooperation is greatly appreciated.
Conducting large-scale field experiments 260 miles from home base is
a difficult undertaking. Besides the cooperation acknowledged above, due
credit must be given to Mr. Robert Ingvalson, Soil Scientist, ARS, who
has provided competent leadership to the small group stationed at Tacna,
Arizona, for the daily operation of the project.
xv
-------
SECTION 1
INTRODUCTION
The salt load of the Colorado River is a major national problem. Unless
corrective actions are taken, the present average salt concentration of about
850 mg/£ in the lower reaches of the river may increase to about 1200 mg/£
by the year 2000 (Bessler and Maletic, 1975). Such an increase would have
serious economic consequences for the seven states adjoining the river and
for the Republic of Mexico. Drainage from irrigated agriculture has been
identified by the U.S. Environmental Protection Agency (1971) as the major
controllable source of salinity in the Colorado River. Thus, it appears that
irrigated agriculture must bear a large part of the burden of reversing the
trend of increasing salinity. Salinity problems, however, are by no means
restricted to the Colorado River.
Research at the U.S. Salinity Laboratory indicates that in many instances
the salt in drain water from irrigated agriculture can be reduced (van Schilf-
gaarde et al., 1974). The key new finding is that if crops are irrigated so
that water of low salinity is available in the upper portion of the root zone,
the soil solution in the lower portion of the root zone can be permitted to
concentrate considerably more than had previously been thought possible with-
out decreasing yields (Bernstein and Francois, 1973). If these same results
hold true under field conditions, the leaching requirement of most crops grown
with Colorado River water could be reduced below 10%. Less leaching would
result in reduced irrigation diversions and lower drainage rates. In turn,
low leaching would reduce soil mineral dissolution and enhance precipitation
°f gypsum and lime, thereby reducing the salt load of the drain waters
(Rhoades et al., 1974). Where the groundwater is saline or the aquifer
provides a source of salt, a reduction in drainage will reduce the salt
load returned to a stream. Furthermore, the smaller volume of drainage water
may make alternative means of disposal feasible.
Achieving uniformly low leaching requires an irrigation system capable
of overcoming the inherently nonuniform infiltration characteristics of most
fields. For nonuniform fields, the infiltration rate must be controlled by
the irrigation system rather than by the soil. This may be accomplished by
applying water uniformly at a rate sufficiently low to avoid surface ponding
and in amounts sufficiently small to avoid saturating the soil profile.
Frequent irrigation also provides a continuous supply of low-salinity water
in the upper root zone. Limited experience indicates that uniformity of
infiltration approaching 95% can be achieved with some well-designed irriga-
tion systems. Even with such systems, programming irrigation to achieve as
little as 5% leaching offers some challenges. Clearly, irrigation based on
-1-
-------
evapotranspiration measurements alone is not feasible because a 1% error in
estimating evapotranspiration causes a 20% change in leaching if the leaching
target is 5%. Precise irrigation management in salt-affected soils requires
knowledge of drain-water flux, measured directly or inferred from soil
salinity and irrigation volume (Oster et al., 1976).
Although a reduction in leaching well below levels generally recom-
mended appears theoretically feasible, the principles and components neces-
sary for its achievement have been demonstrated only on a small scale. To
evaluate its potential for alleviating the salinity problem of a major river
basin, field studies were designed to provide key information spanning the
range of conditions found in the Colorado River Basin. One field study,
consisting of two experiments, has been started in the Wellton-Mohawk Irri-
gation and Drainage District of southwestern Arizona (Fig. 1); its irriga-
tion water quality, climate, cropping pattern, and range of soil properties
complement another field study in progress in the Grand Valley of Colorado.1
The first experiment was installed in December 1973 with citrus on coarse-
textured mesa soil, and the second was started in September 1974 with alfalfa
on medium-textured valley soil.
The primary objective of these field studies is to determine the feasi-
bility of reducing the salt output in drain water by reduced leaching while
maintaining crop yield. This objective is to be accomplished by utilizing
(1) uniform and frequent irrigations, (2) recent concepts of crop salt
tolerance, and (3) recent models of salt losses by precipitation and of
salt gains from mineral weathering. Additional objectives are (1) to
determine the components of the water and salt balance quantitatively under
minimum leaching, and (2) to determine the requirements for irrigation sys-
tems to achieve low leaching under field conditions.
This constitutes an interim report of results obtained to date and
should, therefore, not be considered conclusive. Field data collection and
analyses will continue for 2 years. At that time, the final conclusions and
recommendations will be reported. The additional time is required for sev-
eral years of data collection after the soil salinity profiles have reached
equilibrium.
X1974 and 1975 annual progress reports on "Alleviation of salt load
in irrigation water return flow of the Upper Colorado River Basin" prepared
by E. G. Kruse for the Bureau of Reclamation under Contract No. 14-06-40-
0-5942.
-2-
-------
I
U)
Figure 1. Location of citrus and alfalfa minimum leaching experiments in southwestern Arizona.
-------
SECTION 2
SUMMARY AND CONCLUSIONS
To investigate the potential of reducing the salt load in return flow
from irrigated areas by reduced leaching, two field projects were estab-
lished near Tacna, Arizona in the Wellton-Mohawk Irrigation and Drainage
District. Colorado River water with about 944 mg/A total dissolved solids
was used for irrigation.
In one project, trickle irrigation was used to control the amount of
water applied to each tree in a mature Valencia orange orchard located on
Dateland fine sandy loam. Three treatments, replicated nine times, intended
to apply 5, 10, and 20% leaching water, were established and compared with
conventional border flood irrigation. Three years' results indicate that
the actual leaching percentages obtained were 8, 11, and 22, compared to
47 on the border flood check. The best estimate for annual evapotranspira-
tion is 1400 mm. To date, no differences have been observed in the quan-
tity or quality of fruit among treatments, or between treatments and the
check plots. Soil salinity has increased as anticipated.
If the citrus yields can be maintained over time, the data may be used
to project, for the 3000 ha of citrus now grown in the Wellton-Mohawk
Irrigation and Drainage District, that decreasing the leaching percentage
from 47 to 20 would reduce drainage water by 43.7 x 106 m3/yr and salt load
of this drainage water 45,500 Mg/yr. The short-term effect of this manage-
ment change, however, could be substantially greater. Since the drainage
water currently pumped out of the District has a concentration of about
3000 mg/Jl, the reduction in salt load would initially be 130,000 Mg/yr.
In the other project, controlled low leaching rates were evaluated on
alfalfa grown in Indio fine sandy loam soil. About 2 ha of an 8-ha field
was divided into 15 plots, providing 5 replications of 3 treatments. The
treatments imposed were expected to result in 5, 10, and 20% leaching. A
moving boom-spray irrigation system that applied 6 mm of water each pass
was used. The remainder of the field was irrigated by level-basin flooding.
The actual leaching percentages obtained to date were lower than those
planned, and were probably about 3, 5, and 10. The flooded field received
the same amount of water as the 20% treatment.
The high frequency of irrigation caused continuously wet soil surface
conditions that increased weed growth and, combined with the heavy harvest-
ing equipment, reduced the alfalfa stand. Alfalfa yields on the sprinkled
plots were about 16% less than those on the flooded field.
-4-
-------
Analyses of leachate volumes and concentrations, in situ salinity
sensor readings, and records of water applied provide reasonably consistent
results. They imply that the sprinkled plots were underirrigated, and that
the flooded field obtained some of its water from the rather shallow water
table. The annual ET for alfalfa at the site probably is about 2000 mm.
The salinity profiles have developed rather slowly and probably haven't
reached equilibrium. The salinity is substantially lower in the flooded
field than in the sprinkled plots.
Reseeding of the alfalfa and redesign of the irrigation system so that
fewer passes are required are expected to resolve the management difficul-
ties encountered. It is unlikely, however, that the low leaching obtained
on the flooded field can be substantially decreased further.
-5-
-------
SECTION 3
CITRUS
EXPERIMENTAL PROCEDURE
Experimental Design
The citrus experiment, located 3 km east of Tacna, Arizona, on the
Desert Valencia Ranch, consists of nearly 2 ha of trees centered within a
4-ha block planted in the fall of 1963. The experiment is surrounded by
similar-age trees, except on the south, where it is bordered by the Mohawk
Irrigation Canal. The 4-ha block was chosen based on its history of high
yield and the experimental site was located within the block on the basis of
the uniformity of tree trunk circumference. The Valencia orange trees
(Citrus sinensis L.) are Campbell Nucellar budwood grafted on Rough Lemon
root stock. Tree spacing is 4.9 by 6.7 m. The experimental design is
illustrated in Fig. 2. The randomized block experiment consists of three
treatments of 5, 10, and 20% leaching. Each treatment consists of nine rep-
lications of nine trees each in three- by three-tree plots. Based on the
analysis of Jones, Embleton, and Cree (1957), this experimental design
should permit the statistical detection of 12% yield differences at the 5%
level of significance. The experiment is separated from the remainder of
the grove by border trees irrigated to achieve 20% leaching.
Soil Properties
Three soil profiles within the experimental site were examined by Soil
Conservation Service personnel for characterization and description of the
soil morphological properties. A soil description from samples taken by the
SCS near the center of the experiment is given in Table A-l of the Appendix.
The soil is classified as Dateland fine sandy loam (Typic Haplargid, coarse-
loamy, mixed, hyperthermic) and is representative of the soils where citrus
is grown in the district. The soil is calcareous throughout, well drained,
and moderately permeable. It is underlain with sand beginning at a depth of
1.5 to 2.0 m and continuing to at least a depth of 4 m. Over 300 soil sam-
ples were taken in plots L7, M8, and H6 during December 1973 for chemical
analysis and characterization of initial soil conditions. Sampling tra-
verses were made under the tree canopies in the four major compass direc-
tions. Samples were collected along each traverse at distances of 1.0, 1.4,
1.8, 2.1, 2.4, and 3.0 m from the tree trunk and at depth intervals of 0 to
0.15, 0.15 to 0.30, 0.3 to 0.6, 0.6 to 0.9, and 0.9 to 1.2 m. The samples
were mixed and sieved, then analyzed for cation-exchange capacity, exchange-
able sodium, sodium-adsorption ratio, saturation percentage, field-water
-6-
-------
LEGEND
X Vacuum Extractor
Ml Plot Identification
L 5% Leaching
M 10% Leaching
H 20% Leaching
Border Drip
!o o
I | Border Bubbler
Border Dike
To'
10
!°
H°
|0
10
i:
10
!°
!°
n°
to
"0
io
|0
||0
io°
£
||0
10
!°
s
j|0
10
i!°
1 Io
i!°
!i9.
r°
o
o_
o[
o°i
oj
°!
°o!
0|
Ol
ol
°!
0
o'
°l
0|
01
01
0|
°l
°l
01
01
ol
o1
o[
o
o
0
o
OOOOOOOOOI3
OOOOOOOOO 3
OOOOOOOOOj}
°M°.°
O 0 O
O 0 O
oto2o
O 0 0
O O 0
0L°3°
O O 0
O 0 0
°H°4°
o o o
o o o
0H°5°
oHo5o
o o o
o o o
°M06°
000
O O 0
O O 0
oMo7o
°H°8°
O O O
o o o
0L°9°
O O O
o o o
000
000
o o o
°H°2°
000
o o o
°M°30
0 O 0
o o o
oLo4o
o o o
o o o
0M°5°
oMo5o
o o o
o^o o
0H°6°
oVo
o~o o
00 0
oLo7o
S#S
0*0 o
°H°90
o o o
o o o
°H°,° :
o o o :
0 0 O 3
°M02° D
0003
0 0 0 3
0H°3° ^
o o o :
o o o D
°M°4° 3
O O O 3
O O O D
000 >
ooo :
O O O 3
o o o :
000 5
0 0 0 C
ooo:
ooo:
0H070 3
000 3
0L080 3
O O 0 3
°M°9° 3
O O O G
OOO)
OOOOOOOOOI3
OOOOOOOOO C
ooooooooo :
O O O OO OO O Oj 3
:
Figure 2. Design of minimum leaching experiment on mature
Valencia orange trees in southwestern Arizona.
-7-
-------
content, electrical conductivity, pH, and the following soluble constituents
in saturation extracts: calcium, magnesium, sodium, potassium, bicarbonate,
sulfate, chloride, and nitrate. A similar number of samples will be taken
at the conclusion of the experiment.
The relationship between hydraulic conductivity and matric potential
for Dateland fine sandy loam soil was determined in a temperature-controlled
laboratory from three 0.9-m-deep, undisturbed soil columns taken in the
winter of 1973. Water was applied to the top of the sealed columns by a
constant flow device while a vacuum pump was used to maintain a constant
suction on the drainage collection chamber of each column. Miniature ten-
siometers, spaced 0.1 m apart along the length of the columns, were used to
measure soil matric potential. The columns were operated for about 9
months. Each time steady state was reached at preset inflow rates, soil
matric potentials were recorded and the hydraulic conductivity was calcu-
lated at each tensiometer depth.
Irrigation
The irrigation water is diverted from the Colorado River at Imperial Dam
and delivered to the district in open, concrete-lined canals. The typical
concentration of total dissolved solids in the Mohawk Canal during 1975 was
944 mg/£ and the major salt constituents and their concentrations were Ca,
4.5; Mg, 2.7; Na, 6.8; HC03> 2.8; SO^, 7.7; and Cl, 3.5 meq/£2. Irrigation
water for the experiment is pumped directly from the Mohawk Canal, passed
through commercial sand and screen filters, and delivered in buried, plastic
mains to each plot at a pressure of about 350 kPa. The filters provide ef-
fective filtration of foreign material down to 75 micrometers in size. The
11-kW centrifugal irrigation pump can deliver 16 £/s.
The frequency of irrigation and volume of water applied is controlled
by programmable time clocks which operate the pump and automatic irrigation
controls at each plot. Initially, the time clock for the experimental plots
was programmed to operate the pump for 30 min three (winter) to six (summer)
times every day. During the summer of 1976, the pump operation time was
decreased to 15 minutes and the maximum number of irrigations per day was
doubled. These changes were made to prevent surface ponding under some
trees. A flow control valve at each tree delivers 32 ml/s to each tree
whenever a plot is irrigated. This irrigation volume is equivalent to a
uniform application of 0.9 mm over the entire surface area for a 15-minute
irrigation period. The volume of water applied to each plot of nine trees
is measured by two water meters placed in series to insure accurate, fail-
2The 944 mg/liter reported is the sum of the analytically determined
dissolved solutes. The equivalent concentration in terms of total dissolved
solids is 857 mg/liter. For this conversion of soluble to residue constit-
uents, it was assumed that all constituents would result in anhydrous salts
upon evaporation and that all bicarbonate in solution would exist as carbonate
in the residue. The bicarbonate in solution was divided by 2.03 to determine
its equivalent weight as carbonate in the residue.
-8-
-------
safe measurements. Each of the 243 experimental trees is irrigated with a
35-m spiral of dual-chamber, drip-irrigation tubing as shown in Figs. 3 and
4. The tubing has 0.5-mm-diameter outlets every 0.3 m along its length.
This design was chosen to enhance uniform water application under each tree.
The border trees are irrigated by a second programmable time clock that
operates the pump and automatic irrigation controls for 45 minutes as many as
two (winter) to four (summer) times daily during times when the experimental
plots are not being irrigated. The two border rows along the west side of
the experiment and trees bordering the experiment on the north and south are
irrigated by bubblers (see Fig. 4) filling small basins formed under each
tree. The water delivery rate of each bubbler is controlled at 63 ml/s by
a flow control valve. The two border rows along the east side of the ex-
periment and the closely spaced trees on the north edge of the block are
irrigated from capillary tubes (1.7 mm ID) inserted inside 25-mm-diameter
polyethylene pipe (see Fig. 4). The pipe was laid on the soil surface
about 1 m out from the trunk on both sides of the trees. The capillary tube
outlets are spaced 0.7 m apart, giving 14 emitters per tree. Flow control
valves in each pipeline deliver an average of 32 ml/s to each tree.
Instrumentation
Each plot is irrigated automatically based on the readings of four ten-
siometers installed at the 0.3-m depth and located 60 degrees apart 1.5 m
radially out from the trunk of the center tree of each plot (Fig. 3). To
sense soil matric potential, a light-emitting diode and a phototransistor
are placed directly opposite each other on the manometer columns of the tensi-
ometers (Austin and Rawlins, 1977). As the soil dries, rising mercury in the
manometer column interrupts the light beam from the diode. This electrically
opens a water valve to irrigate the plot whenever the pump is operated. To
help overcome variability and to guard against tensiometer failure, the elec-
trical signal to open the valve must come from any two of the four tensiom-
eters. Irrigation frequency is controlled by the location (setpoint) of
the phototransistor on the manometer column. Lowering the setpoint raises
the soil matric potential and increases leaching; raising the setpoint de-
creases leaching.
Salinity sensors (Richards, 1966; Oster and Willardson, 1971) and addi-
tional tensiometers in two spatial distributions were installed beneath the
center tree of each plot. For three of the nine replications for each
leaching treatment, four sets of salinity sensors and two sets of tensiom-
eters were installed along one radial (see Fig. 3). All the tensiometers
and the salinity sensors at'0.15-, 0.30-, and 0.45-m depths were installed
initially, and those at 0.60- and 0.90-m depths were installed in June 1974.
Initially, the salinity sensors were installed at relatively shallow soil
depths because of the expected time lag in salinity buildup with depth and
the need for responsive salinity feedback to control irrigation. During
October and November 1974, the soil salinity at 0.90 m began to increase.
In addition, the soil salinity at the 0.15-m depth equaled or exceeded
that at 0.30 m and also tended to fluctuate more rapidly than at 0.30 m.
Because the leaching percentage and soil salinity at the bottom of the root
zone are of primary interest, the sensors buried at a depth of 0.15 m were
9
-------
Tree Drip Line
Duol Chamber
Trickle
Tubing
35m. long
Scale', f-
I m
Tensiometers
Salinity Sensors'.
For Irrigation, 0.3 - m deep
A Depths of 0.15,0.30, 0.45,0.60,0.90m
jg Depths of 0.30, 0.45, 0.90, 1.50 m
(V) Depths of 0.30. 0.45,0.60, \-ZO m
Figure 3. Illustration of instrumentation under center tree of three of the
nine replications for each leaching treatment. The remaining replications
have two sets of salinity sensors and one set of tensiometers.
-10-
-------
Dual - Chamber Irrigation Tubing
Distribution Orifices
(Emits water at law pressure)
Supply Orifice
(Feeds water to distribution tube)
Main Chamber
(Water supply tube)
Secondary Chamber'
(Water distribution tube)
Drip Irrigation Tubing
-25 mm Polyethylene
Tubing
Bubbler Irrigation System
Flow Control Valve
Figure 4. Sketch of dual-chamber, bubbler, and drip irrigation systems.
-11-
-------
reinstalled at depths of either 1.2 or 1.5 m in March 1975. The present
depth pattern of salinity sensors for these three replications is four
sensors at depths of 0.3 and 0.45 m and two at depths of 0.6, 0.9, 1.2, and
1.5 m. For the remaining six replications, only two sets of salinity
sensors and one set of tensiometers were installed. The salinity sensors,
at depths of 0.30 and 0.45 m, were located 1.25 and 1.65 m out from the
trunk on the same radial line and the tensiometers, at the depths given in
Fig. 3, were installed midway between the salt sensor sets. The salinity
sensors and tensiometers are read twice weekly.
To obtain detailed information on the instantaneous water flow pattern
under a citrus tree, 84 tensiometers were installed under the center tree
of plot H4 in March 1975. The tensiometers, identical to those installed
in December 1973, were installed at depths ranging from 0.3 to 1.8 m and at
0.6-m intervals on three radial lines out from the tree trunk; one radial
line was in the tree row, one was perpendicular to the row, and the third
was diagonal.
Four vacuum extractors were installed in one replication of each of the
three leaching treatments to measure the leaching rate and the chemical
composition of the soil solution below the undisturbed root zone (Duke and
Haise, 1973). Within a treatment, extractors are directed toward the center
of four different trees from a common manhole. A diagram of the extractor
is given in Fig. 5; note location in Fig. 2. Each extractor consists of
three independent sheet-metal troughs (0.15 m wide by 0.20 m high by 0.61 m
long). The extractors were installed in a rectangular tunnel formed by
first augering a horizontal hole at a soil depth of 1.2 m and then forcing
a rectangular shaper into the hole. Each extractor contained two lines of
ceramic tubes 12 mm in diameter and was filled with soil removed in forming
the tunnel. The extractors were raised against the smooth ceiling of the
tunnel by inflating a butyl rubber air pillow. Soil solution is collected
in a small sample bottle in each extractor drain line just ahead of a large
collection bottle held under partial vacuum. With sufficient flow, the
contents of the small sample bottles are continually replaced by fresh soil
water, providing samples in equilibrium with the partial pressure of the
soil CO-. The vacuum is adjusted so that two tensiometers over the center
of the extractor read the same as two tensiometers about 0.3 m away from
the extractor at the same depth (see inset in Fig. 5). With uniform soil
matric potential near their tops, the extractors should intercept the flux
representative of their cross-sectional area without causing convergence
or divergence of flow.
Soil Chloride Distribution
Soil samples are taken annually to determine the chloride distributions
under selected trees. The chloride data are useful in locating the depth of
the root zone and indicating water uptake and salinity distributions, and
providing complementary data to the salinity sensor readings. The water up-
take distribution is required in reassessing the anticipated soil salinity
at shallow depths for feedback in controlling irrigation. The maximum
chloride concentrations, which are indicative of the lower boundary of the
root zone, provide an estimate of leaching fraction. The chloride distri-
-12-
-------
CROSS-SECTION
A A
DUAL CHAMBER
TRICKLE TUBING
N,
SHEET METAL
TROUGH
Im.
Figure 5. Schematic of vacuum extractor installation.
-------
bution data, in conjunction with the amounts of applied water and the
chloride concentration in the leachates, provide another technique of
estimating evapotranspiration. The chloride data also provide a measure
of the variability of soil salinity.
In December 1974, the average chloride distribution with soil depth was
determined by sampling beneath three trees in three of the 20% (H3, H4, and
H6) and three of the 5% (L3, L5, and L7) leaching plots. The sampling sites
were about 1.2 m from the tree trunk at three randomly selected locations
around each tree. The soil samples were divided in 50-mm increments to a
depth of 100 mm and in 100-mm increments between 100 and 1800 mm. The areas
beneath the center trees of plots H4 and L7 were also sampled in February
1975 and March 1976. The sampling pattern for 1975 and 1976 is illustrated
in Fig. 6. The chloride concentrations were expressed in milliequivalents
per liter at field water content at the time of sampling.
Irrigation Management
The ultimate objective of the experiment requires that leaching at the
bottom of the root zone be controlled precisely. Since routine measurements
of water flux below the root zone are not feasible, our management scheme
was to control the salinity at the bottom of the root zone at a level corre-
sponding to that predicted for the imposed leaching fraction. Because of
the lag between changes in water application and corresponding changes in
soil salinity with depth, the irrigation system is controlled directly by the
irrigation tensiometers described previously. Irrigation is applied when-
ever the soil matric potential as measured by these tensiometers decreases
below a predetermined setpoint. Control of matric potential can assure ade-
quate water for plant growth, but it will not assure a prescribed leaching
percentage, because the relationship between hydraulic conductivity and
matric potential is too variable and because the depth of rooting, and thus
the percentage of water lost at the tensiometer depth, is not known, and in
any case, may vary. Thus, a salinity measure is needed as feedback to the
irrigation control. To overcome the long lag between water application and
soil salinity changes at depth, salinity sensor readings at the 0.3-m soil
depth are used as feedback.
The initial values of soil salinity at the 0.3-m soil depth used for
control were estimated from salinity distributions calculated by means of the
soil water composition model of Oster and Rhoades (1975). These distribu-
tions depend on the chemical composition of the irrigation water, the partial
pressure of C0_ in the soil, and the water uptake pattern of the crop, as
well as the leaching percentage. Using the water uptake data for citrus re-
ported by Erie, French, and Harris (1965) and a typical Colorado River water
composition, we calculated the target distributions given in Fig. 7 for
February 1974. These calculations indicated target salinities of 0.18, 0.20,
and 0.22 S/m for the 20, 10, and 5% leaching treatments at the 0.3-m depth,
with corresponding values at the bottom of the root zone of 0.5, 0.9, and
1.3 S/m, respectively. The tensiometer setpoints were chosen initially on
the basis of best judgment. Decisions on adjustment of the tensiometer
setpoints for each plot were made based on biweekly evaluation of the soil
salinity, the soil matric potential profile, and the amount of irrigation
-14-
-------
x = Feb '75
76
X X
DRIP
AREA
Figure 6. Sampling pattern for soil chloride determinations beneath the cen-
ter tree of plots L7 and H4 in 1975 and 1976.
-15-
-------
Soil Water Salinity , S/m
0 0.2 0.4 0.6 0.8 1.0 1.2 i.4 1.6
Figure 7. Computed values of electrical conductivity with soil depth for Valencia orange trees irri-
gated with Colorado River water at 5, 10, and 20% leaching.
-------
water applied. As knowledge was gained from soil chloride distribution,
leachate volumes and salt concentrations from the vacuum extractors, soil
salinity deeper in the profile, and the partial pressure of CO- in the soil,
the target salinity levels at the 0.3-m soil depth were also changed.
Table 1 lists the projected salinity values used as feedback in controlling
irrigation. The current projected soil salinity distributions as a function
of leaching are also given in Fig. 7.
TABLE 1. ALTERATIONS OF THE PROJECTED ELECTRICAL CONDUCTIVITY OF THE SOIL
WATER (S/m) AT THE 0.3-m DEPTH TO ACHIEVE THE DESIRED LEACHING IN
THE CITRUS EXPERIMENT
Date Leaching Treatment
5% 10% 20%
Dec.
Aug.
Jan.
Mar.
Sep.
1973
1974
1975
1975
1976
0.
0.
0.
0.
0.
22
34
29
28
36
0.
0.
0.
0.
0.
20
29
26
25
29
0.
0.
0.
0.
0.
18
24
21
22
22
Four irrigation tensiometers were also installed under one tree for
each of the two borders. Since the water application geometry for the
borders is different from that of the experimental plots, the tensiometer
setpoints are adjusted so that the same amount of water is applied to the
borders as the average application for the 20% leaching treatment. Salin-
ity sensors are not used as feedback to control irrigation for the borders.
Check Plots
The citrus grove on the ranch is irrigated by border flooding; a border
consists of 6 rows of 35 trees each, a total area of 0.7 ha, surrounded by
earthen dikes. A typical border irrigation consists of a 150-mm-deep appli-
cation over the entire area within the border. The water is applied to each
border in about 45 minutes through six sliding gates in the concrete wall of
the irrigation lateral. Such a border was selected for comparison in the
4-ha block of trees immediately north of the experiment. Soil matric poten-
tial is monitored under three separate trees with tensiometers as described
above. Because of the high leaching percentage, the soil salinity level is
very low and is monitored periodically from soil samples. For yield and
quality comparison, 27 trees from the center of three rows in the middle of
the border are harvested individually. Water applied is measured with a con-
crete, critical-flow flume (Replogle, 1977) installed in the irrigation
lateral and elapsed-time devices installed on the irrigation outlets. Water
flow in the lateral can be calculated to within about 2% from the known
-17-
-------
flume geometry and by continuously recording water elevation at the entrance
relative to the flume floor at its throat.
One border of flood-irrigated trees just east of our experiment within
the same 4-ha block is fertilized by foliar application, as are the experi-
mental trees, rather than through the irrigation water as is done throughout
the grove. This border is also used for yield comparisons with the experi-
mental trees.
Yield and Fruit Quality
The fruit from all the experimental trees, the two rows of border trees
parallel with the experimental trees on the east and west, and three rows of
trees from both the flood- and fertilizer-check plots are harvested by indi-
vidual tree in April of each year and weighed. The number of fruit per tree
is determined by counting the number of fruit in one field box from each tree
and calculating the total number from the average weight per fruit.
Just before harvest, four fruit are picked from each experimental tree
for quality analysis. Three samples of 12 fruit each are also picked from
the borders and the flood- and fertilizer-check plots. Measures of fruit
quality include: fruit length and width, rind color and texture, and ring
size. Ring size is equivalent to the average number of fruit contained in a
standard-size shipping carton. After the fruit analyses, the juice is
extracted and analyzed. Juice analyses include total soluble solids, total
acid, and percent juice.
Tree Growth and Leaf Analysis
A simple measure of tree growth often used in citrus research is trunk
enlargement with time. Although the trees are over 10 yrs old, they are not
fully mature, and trunk circumference is measured annually.
Leaf samples were taken initially in March 1974, and yearly in Septem-
ber thereafter, to evaluate nutritional status and detect possible toxic
levels of chloride and sodium. Each sample consists of 72 leaves obtained
by sampling eight leaves from each of the nine trees per plot. Leaves that
represent average foliar conditions are taken from the ends of nonfruiting
terminal branches at two heights on four sides of each tree. Leaf samples
are handled, cleaned, and prepared for analysis according to established pro-
cedures (Reisenauer, 1976). The leaf samples are analyzed for 13 mineral
elements.
Agronomic Practices
Except for fertilization, all crop management operations such as frost
protection, pruning, and insect and weed control are performed by the ranch
to match those given the surrounding groves. The experimental trees are
fertilized with foliar sprays of urea and microelement chelates. Foliar
application was chosen to permit uniform fertilizer applications to each
tree while differential water applications were made, to minimize nitrate
discharge in the leachate, to avoid further salt additions to the irrigation
-18-
-------
water, and to minimize the amount of fertilizer required. Timed applications
of nitrogen as low-biuret urea are applied to each tree six times each spring
at a rate of 115 g N per tree per application. Chelates of iron, zinc, and
manganese are also applied with some nitrogen applications at the rates of
1.0, 1.0, and 0.5 g per tree per year, respectively.
Soil Air
With frequent irrigations to maintain the soil matric potential near
-10 kPa, soil aeration may be a problem, particularly in heavy soils. Even
though no aeration problems were expected in the sandy loam soil of the
citrus experiment, soil oxygen concentrations were measured during both the
winter and summer of 1975, in cooperation with Dr. Burl Meek of the Imperial
Valley Conservation Research Center in Brawley, California, with polari-
graphic probes installed at the 0.45-m soil depth and about 1.5 m radially
from the tree trunk. Because the carbon dioxide concentrations in the soil-
air phase will affect the chemical composition of the salt load of the
drainage waters, soil-air samples were obtained by attaching evacuated
plastic bags to porous aeration stones buried in the soil and the carbon
dioxide content was determined by using an infrared gas analyzer.
RESULTS
Soil Properties
Analyses of the initial soil samples, summarized in Table 2, show that
soil salinity was generally low at the beginning of the experiment, but a few
moderately salinized zones of apparently restricted permeability were found.
Chloride concentrations in the soil samples and distributions in the soil
profile suggest that past flood irrigation management has resulted in a
leaching percentage of about 40 at the 1.2-m soil depth. A few profiles
showed evidence of leaching as low as 8%; these profiles correspond to areas
with relatively high clay contents near the soil surface.
The relationship between hydraulic conductivity and matric potential
for Dateland fine sandy loam soil is given in Fig. 8. These data were ob-
tained from three 0.9-m-deep undisturbed soil columns taken in the fall of
1973. The horizontal lines in the figure indicate the variability in each
measured point.
Water Use
The average depth of water applied daily for the three leaching treat-
ments is given in Fig. 9 by months for the period of February 1974 to
September 1976. The rate of application on the flood check plot is given
from January 1975 to September 1976. These water application rates are the
sum of irrigation and rainfall and are calculated for the total area allo-
cated each tree, 32.7 m2. The average number of liters applied daily to
each tree in a given treatment may be calculated by multiplying the depth
given in Fig. 9 by 32.7.
-19-
-------
Matric Potential , kPa
-20 -10
0
5000
Figure 8. Relationship between soil matric potential and hydraulic conductiv-
ity for Dateland fine sandy loam soil.
-20-
-------
I
to
^
£ E
o
tr
CO
§
o
o>
c
o
o
o
Q.
<
O
a.
ai
o
16
I 5
14
13
12
I I
10
9
8
7
6
5
4
3
2
I
0
JFMAMJJASONDJFMAMJJASONDJFMAMJJAS
1974 1975 1976
Figure 9. Average daily water application for the three leaching treatments and the flood check by
month from February 1974 to September 1976.
-------
TABLE 2. REPRESENTATIVE SOIL PROPERTIES OF DATELAND FINE SANDY LOAM SOIL
Unit Typical Values Range
Cation-exchange capacity meq/lOOg
Exchangeable-sodium percentage %
Sodium-adsorption ratio
Saturation percentage %
Field water content
EC
PH*
a.
Soluble"
It
It
"
II
II
It
II
calcium
magnesium
sodium
potassium
bicarbonate
sulfate
chloride
nitrate
g/g
S/m@25°C
meq/1
"
it
it
it
ti
ti
it
9
3
3
25
11
0.
3
5
5
2
- 13
- 4
- 4
- 30
- 12
13-
7.4
-
2
-
0.2
2
-
-
0.6
0.15
4
7
6
3
2.
3
3
19
4
1
6.
1.
0.
3
tr
0.
2.
0.
0.
5
4
6
7
8
3
9
2
- 23
- 13
- 10
- 50
- 22
- 6
- 8.
- 23
- 11
- 37
- 0.
- 4.
- 56
- 28
- 6.
3
6
6
2
*
Soluble in saturation extract.
The annual water application amounts for the three leaching treatments
are given in Table 3, along with those for the two borders and the flood
check plot. For comparison, annual pan evaporation and rainfall are also
given in Table 3. The evaporation pan occupies the area normally taken by
a citrus tree in a grove about 0.4 km from the experiment. Pan evaporation
values from an open area about 5 km from the grove are given in the alfalfa
section. Pan evaporation from the citrus grove has been about 85% of that
from an open area.
Water was withheld in the beginning to increase the soil salinity to
the projected levels for the three leaching treatments; thus, the total
amount of water applied in 1974 was no doubt too low to be typical. On the
other hand, based on the relationship between pan evaporation and calculated
evapotranspiration shown in Fig. 10, we realized that our irrigation manage-
ment scheme lagged changes in pan evaporation. Thus, in the summer of 1976,
we attempted to anticipate pan evaporation; unfortunately, we overcompen-
sated and applied too much water in the summer of 1976. As a consequence,
either the water application data for 1975 or the 3-year average is used as
our current estimate of the water requirements for the three leaching treat-
ments. Our attempts to achieve 20% leaching on the two borders, based on the
average amount applied to the 20% leaching treatment, were successful. About
50% more water was applied to the flood check plot than to our experimental
trees.
Evapotranspiration (ET) for these trees can be estimated by multiplying
the depth of water applied by 0.95, 0.90, and 0.80 to account for the de-
sired leaching percentages of 5, 10, and 20. This estimate of ET does not
account for failing to achieve the desired leaching percentage or for
-22-
-------
TABLE 3. ANNUAL IRRIGATION, RAINFALL, AND PAN EVAPORATION FOR
VALENCIA ORANGE TREES
Year
Depth of water applied* (nan)
Leaching Treatment
1974
1975
1976*
Average
5%
1401
1498
1675
1525
10%
1450
1560
1725
1578
20%
1685
1802
1975
1821
Borders
Bubbler
1651
1942
2049
1881
Drip
1818
1912
1704
1811
Flood
Check
2582
2750
2666
Rain-
fall
nun
105
84
90
93
Pan Evapo- Cal-
ration culated
mm ET (mm)
1838
1732
1650
1740
1325
1450
1550
1440
*Irrigation plus rainfall.
^Data estimated for October, November, and December.
changes in soil water storage. The actual ET for each leaching treatment
should be the same, unless growth or stomatal aperture is influenced by
these small differences in leaching. ET has been assumed to be independent
of the leaching treatments imposed, because data presented below indicate
no significant differences in trunk circumference. The annual estimates of
ET presented in Table 3 show the 3-year average as 1440 mm.
As an independent check of the ET estimate based on water applied, ET
was computed by the modified Penman equation (Doorenbos and Pruitt, 1975)
and the Jensen-Haise equation (Jensen, 1973). The potential ET calculated
from these two equations was multiplied by a crop coefficient, varying from
0.5 in winter to 0.6 in summer, to obtain ET. The results of these computa-
tions of ET are given in Table 4 as average daily ET rates by month. Also
listed in the table are the estimates of ET calculated from water applica-
tion and the consumptive use of mature Navel orange trees near Phoenix,
Arizona, published by Erie, French, and Harris (1965). ET calculated by the
modified Penman equation agrees well with the ET calculated in our experi-
ments.
The average annual ET based on the modified Penman and Jensen-Haise
equations are 1421 and 1342 mm, respectively, compared to 1440 mm based on
water applied. The data of Erie et al. (1965) were consistently lower, with
the average annual ET being 1000 mm.
The average daily ET based on water applied and computed by month was
compared to the average daily rate of pan evaporation in the grove in Fig.
10. As expected, the shapes of the curves are similar, but the calculated
ET lags pan evaporation throughout. This lag is no doubt caused by a lag in
the irrigation management. Based on these data, the pan factor (ratio of ET
to pan evaporation) is 0.84.
-23-
-------
O
T3
E
E
CL
i
o
Q
Evapotranspiration
calculated from
amount of Water
applied
JFMAMJJASOND
Month of the Year
Figure 10. Comparison of the daily rate of pan evaporation and evapotranspi-
ration estimated from the amount of water applied for Valencia orange
trees in southwestern Arizona. Data averaged by
month from January 1974 to September 1976.
-24-
-------
I
NS
TABLE 4. COMPARISON OF VARIOUS ESTIMATES OF EVAPOTRANSPIRATION OF VALENCIA
ORANGE TREES IN SOUTHWESTERN ARIZONA
Month
Erie*
et al.
Modified
1974 1975
mm/day
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Yearly
(mm)
1.1
1.6
1.8
2.4
2.9
3.9
4.3
4.3
3.9
2.9
2.1
1.1
total
1000
1.2
2.3
3.2
4.4
5.5
6.6
6.2
6.2
5.2
3.2
1.8
1.4
1438
Penmant
1976 Avg.
mm/ day
1.5
2.1
3.1
3.9
5.4
6.5
6.8
6.5
4.8
3.3
2.0
1.1
1434
1.1
2.0
3.3
4.2
5.4
6.4
6.3
5.8
-
-
-
-
-
1.3
2.1
3.2
4.2
5.4
6.5
6.4
6.2
5.0
3.2
1.9
1.2
1421
Jensen-Haise §
1974 1975 1976
mm/day
1.2
1.8
2.9
3.9
5.1
6.4
6.0
6.2
5.0
3.1
1.8
1.1
1357
1.2
1.6
2.5
3.4
4.9
6.2
6.5
6.4
4.9
3.2
1.8
1.1
1334
1.1
1.7
2.8
3.7
4.9
6.3
6.2
5.8
-
-
-
-
-
Avs.
1.2
1.7
2.7
3.7
5.0
6.3
6.2
6.1
5.0
3.2
1.8
1.1
1342
Calculated from expt.
1974 1975 1976 Avs.
mm/ day
0.9
1.3
1.9
2.8
4.2
6.2
6.4
6.6
6.9
3.4
1.4
1.1
1325
0.9
1.3
2.2
4.0
5.2
6.1
7.0
6.9
6.2
3.5
1.9
1.4
1450
1.2
1.4
2.0
3.3
5.3
8.1
10.8
9.0
5.3
-
-
-
-
1.0
1.3
2.0
3.4
4.9
6.8
8.1
7.5
6.1
3.4
1.6
1.2
1440
Data taken from Erie et al. (1965) .
Calculated by modified Penman equation described by Doorenbos and Pruitt
meteorological information taken by the Irrigation Management Service of
tion, Wellton, Arizona.
n
Calculated by the Jensen-Haise equation described by Jensen (1973).
(1975). Data based on
the U.S. Bureau of Reclama-
-------
Soil Salinity
The time course of average soil salinity measured with salinity sensors
is given in Fig. 11 for the three leaching treatments. (Salinity data for
the 0.6- and 1.2-m soil depths are not given to simplify the figure.) Ini-
tial soil-salinity levels were about equal to the salinity of the irrigation
water, 0.13 S/m, because of previous overirrigation. Beginning in February
1974, soil salinity increased with time until quasi-stable values were at-
tained at the 0.3- to 0.45-m soil depth by July (5 months) and at the 0.9-m
depth by January 1975 (11 months). Since March 1975, changes in soil salin-
ity have undergone two complete cycles. The cyclic pattern is evident to a
soil depth of 0.9 m and the cycle covers about 1 year. During the first
half of the year, the trees are underirrigated and soil salinity increases;
during the last half of the year they are overirrigated and salinity de-
creases. Although cyclic, the differences in salinity among leaching treat-
ments are as expected, with 5% leaching being the most saline and 20% leach-
ing the least.
Fig. 12 compares time-averaged soil salinity as a function of soil
depth for the two time intervals of January 1974 to July 1975 and July 1975
to July 1976, the initial soil-salinity distribution, and the projected
soil-salinity distribution at equilibirum from the model of Oster and
Rhoades (1975). In general, soil salinity has increased with time at all
soil depths and leaching percentages. The soil-salinity distribution is
nearly equal to or exceeds the projected salinity values for all three
leaching treatments to a depth of 1.5 m, except below the 1.2-m depth in
the 5% leaching treatment.
Soil Matric Potential
The tensiometer readings serve primarily as a check on the irrigation
management. Of course, the readings also give the soil matric potential
profiles for the different leaching treatments. The average profiles for
the three leaching treatments during 1974 and 1975 are given in Fig. 13.
As anticipated, the soil matric potentials were very high near the soil
surface, with essentially no differences among treatments. With depth, sig-
nificant differences among treatments did appear and at a depth of 0.9 m,
the soil matric potential averaged -24 kPa for 5% leaching and -14 kPa for
20%.
Soil matric potential was not held steady throughout the year and the
changes that occur can be assessed from Appendix Figs. A-l through A-9,
which show the potential distribution below the center tree in plot H4 at
nine different times as determined from 84 tensiometers. The data are pre-
sented as total head, using the 0.3-m depth as a reference. On January 22,
1976, the plot was irrigated on the basis of a tensiometer setpoint of -9.0
kPa at a depth of 0.3 m. This setpoint more or less fixes the heads at the
0.3-m depth. The hydraulic gradient varied from around 2 between depths of
0.3 and 0.6 m under the wetted area, to 3 farther from the trunk at the
same depth interval, to less than unity at deeper depths. The setpoint was
changed to -8.5 kPa on February 24. On March 23, 1976, the total head was
everywhere lower than it was in January. At deep depths, the gradients
-26-
-------
I
1X3
JFMAMJJASONDJFMAMJJASONDJFMAMJJASOND
1974 1975 1976
Figure 11. Salinity trends with time for the three leaching treatments of the citrus experiment
at various soil depths.
-------
Soil Water Salinity , S/m
0.2 0.4 0.6 0.8 1.0 1.2 1.4
E
.c
a.
a
,
"o
CO
u
0.3
0.6
0.9
1.2
i <5
1 I 1 i
\ \ \\. 5% Leaching
- I vov
V.. \
\ X\ \ x-PROJECTED
'A S SOIL SALINITY
- \ },/
/ "'*.._
0 0.2 0.4 0.6 0.8 1.0 1.2 l.<
0
0.3
0.6
0.9
1.2
1 i i i i I
, . .. 10% Leaching
. 1 ''\X^ JAN '74 (Initial)
\ X '% \ JAN '74 to
\\ \ JULY '75
\ x*'-Sv JULY '75 to
, \N. JULY '76
'/..
0 0.2 0.4 0.6 0.8 1.0 1.2 1.
0.3
0.6
0.9
1.2
i *
_ . v.. 20% Leaching
» V\S
- \ \\
- i tt
- \ 1
\
Figure 12. Time-averaged soil salinity distributions with soil depth for
the initial, two intermediate time periods, and
the projected final conditions.
-28-
-------
-25
Soil Matric Potential, kPo
-15 -10 -5
-20
T
T
5% Leachingix',0%/
.2
.4 .
j=
CL
Q
.6 -
o
en
.8
1.0
Figure 13. Average soil matric potential profiles for the three leaching
treatments in the citrus during 1974 and 1975.
-29-
-------
were near unity. Early in April, the soil salinity began to rise sharply.
On April 28, 1976, the setpoint was still -8.5 kPa, and the total heads were
lower than a month earlier. During April, the ET increased sharply and,
accordingly, the profile should have been permitted to become wetter. This
was not achieved until after the middle of May when the setpoint was
lowered to -6.5 kPa. On August 24, 1976, the setpoint was -5.5 kPa. The
profile was everywhere wetter than it was during the winter and spring and
the hydraulic gradients ranged from 1.5 between depths of 0.3 and 0.6 m, to
near unity at deeper depths.
From 0930 hrs on August 24, 1976 until 2040 hrs on August 29, 1976, the
tree was not irrigated. All tensiometers were read twice daily. By the
morning of August 28, soil drying had caused the -20-kPa contour to shift
upward slightly and the region with total heads between -15 and -20 kPa to
increase substantially. Everywhere the total head was smaller than -10 kPa.
By the evening of August 29, the -20-kPa contour had moved up a little
further and the region with total heads between -15 and -20 kPa had de-
creased at the expense of a region with total heads below -20 kPa in the
0.3- to 0.6-m depth interval. In the afternoon of September 1, 1976, 3 days
after the irrigation had again been turned on, the -20-kPa contour was
closer to the soil surface than it was on August 28; the region with total
heads between -15 and -20 kPa was very small; and the region with total
heads smaller than -20 kPa near the soil surface had disappeared entirely,
except for large distances from the tree on the diagonal radial. At that
time, the gradients in the 0.3- to 0.6-m-depth interval were very large,
while at deeper depths, there was nearly hydrostatic equilibrium. The dis-
tributions on September 4 and 9, 1976, illustrate the return to the poten-
tial distribution prevailing before the water was turned off. The time
course of the total head distribution during the period August 24 to Sep-
tember 1 strongly suggests that there is little root activity below the
0.6-m depth.
Soil Chloride Distribution
After 1 yr, chloride concentrations were highest in the plots that
received the least amount of water, i.e., the 5% leaching plots. The
chloride concentration was maximum in the 0.9- to 1.2-m soil-depth inter-
val, indicating this was the lower boundary of the root zone. This con-
clusion was considered suspect because only 11 months had elapsed since
the start of the experiment. Composite cross sections of the chloride
distribution with soil depth along the three lines sampled in both 1975
and 1976 are shown in Fig. 14 for 5 and 20% leaching. (Individual cross
sections are given in Appendix Figs. A-10 and A-ll.) An overview of the
distributions for both years clearly shows the highest concentrations in
both treatments were beyond the tree canopy. The chloride concentrations
were higher for 5% leaching than for 20% and increased during the time be-
tween the sampling dates in both treatments. Lateral water movement and
water loss beyond the tree canopy are also obvious from the figures.
The chloride distribution for plot L7 indicates that the bottom of
the root zone is saucer shaped. At a distance of 3.3 m from the tree, the
concentration is maximum at a depth of 1.1 m as compared to depths of 1.2
-30-
-------
Distance from Tree , m
FEB '75
3.6 3.0 2.4 1.8 1.2 0.6 0
drip area
0.3-
MAR '76
3.6 3.0 2.4 1.8 1.2 0.6 0
drip area
i
20%
5%
Soil Chloride Distribution
meq/l 0-15 15-30 30-60 60-90 > 90 Cen*err. T.ree0 . ^nt-rtnc
M x = Salinity Sensor Locations
Figure 14. Composite cross sections of soil chloride distribution under the
center tree of a 5% (L7) and 20% (H4) leaching plot after I and 2 years.
-------
and 1.8 m at distances of 3.0 and 0.6 m away from the tree. The calcu-
lated leaching percentages at the bottom of the root zone for plots H4 and
L7 are given in Table 5. The calculated evapotranspiration for plots H4
and L7 is 1211 and 1246 mm, respectively, from calculations based on the
weighted average leaching of 30 and 6% and water applications of 1730 and
1340 mm.
An assessment was made of the nature of variability in soil chlorides,
and, by inference, soil salinity. A log transformation of the chloride
concentrations increased the number of significant sources of variation
(depths, depths x leaching percentage, depths x trees, and leaching treat-
ments) in an analysis of variance. It reduced the coefficient of variation
of the measurement error from 56% for untransformed data to 16%. The data
were also classified into 11 arrays from a total of 681 observations, for
which the arithmetic means were not significantly different at the 5%
level. The standard deviation was independent of the mean based on log-
transformed chloride concentrations (Table 6), which was not true for
untransformed data. Figure 15 shows the frequency distribution of the 681
observations in terms of the variable
where C_, represents the mean chloride concentration of the group and S
represents its standard deviation. The log-transformed data are more
nearly normally distributed. This result is not considered to be unusual by
statisticians, since it is characteristic of dependent variables (C,,-,)
that are inversely related to the independent variable (leaching percent-
age) .
The distribution of water lost from a soil root zone is determined by
surface evaporation, root uptake, and leaching fraction. With saline irriga-
tion water, the fraction of water lost as a function of soil depth can be
calculated based on knowledge of the leaching fraction (LF) , the chloride
concentration of the irrigation water (Cl. = 3.3 meq/£) , and the chloride
concentration of soil water at a given depth (Cl ), since chloride does not
precipitate and negligible quantities are taken up by plant roots. Thus, the
relative water lost (RWL) can be calculated from
RWL = [1 - (Cliw)/(Clsw)]/(l - LF). (2)
The term (1 - LF) is the fraction of the irrigation water that is lost in
the total root zone and (1 - Cl. /Cl ) is the fraction of the irrigation
water lost in the root zone above a given soil depth. At a soil depth where
LF = Cl. /Cl , root water uptake has ceased.
iw sw'
The mean chloride concentrations and relative water loss as functions
of soil depth are given in Table 7 for plots H4 and L7. The Cl concentra-
tions are the means of the seven sites beneath the tree canopy sampled in
March 1976. The leaching percentages for plots H4 and L7 are 30 and 6, re-
spectively. The water loss distribution is similar for both trees, but this
may have been caused by less stable soil salinity levels in plot L7 than in
H4. Transients in soil salinity during the 12 months before soil sampling
-32-
-------
TABLE 5. LEACHING PERCENTAGES FOR TWO PLOTS OF THE CITRUS EXPERIMENT AS
CALCULATED FROM THE CHLORIDE CONCENTRATIONS
IN THE SOIL AND THE IRRIGATION WATER
Distance from tree Plot area represented
(m2)
Calculated leaching percentage
Plot H4 Plot H7
0 -
0.9 -
2.0 -
2.9 -
Weighted
0.9
2.0
2.9
3.3
average
2.54
10.02
11.45
8.69
___
29
37
39
11
30
14
7
5
4
6
TABLE 6. MEAN AND STANDARD DEVIATIONS (S) OF IN. SITU CHLORIDE CONCEN-
TRATIONS FOR ORIGINAL AND Hn TRANSFORMED DATA
Array
Number
of samples
Not transformed
Transformed
Mean
S
Mean
A2
B2
All
D
C2
B13
cm
B121
C112
B112
Bill
81
108
42
177
108
36
57
30
15
15
12
6.0
8.9
9.3
10.6
11.3
13.9
14.6
30.6
31.4
35.5
55.7
meq/H
2.3
5.0
4.3
5.0
6.1
5.7
7.1
8.7
10.2
14.6
20.8
1.73
2.07
2.14
2.26
2.29
2.56
2.58
3.38
3.40
3.49
3.93
.34
.45
.40
.45
.24
.38
.45
.27
.33
.41
.47
-33-
-------
00
OT
.2 80
(-0
X)
O
54CH
-------
TABLE 7. CHLORIDE CONCENTRATION AND WATER LOSS DISTRIBUTION UNDER
CENTER TREES OF 5 AND 20% LEACHING TREATMENT PLOTS
Soil
depth
Plot H4 (20%)
Plot L7 (5%)
Cl
concentration
Accumulated
RWL*
Cl Accumulated
Concentration RWL*
Average
Accumulated
RWL*
m
meq/5,
Relative water loss.
meq/5,
0.3
0.6
0.9
1.2
1.5
1.8
6.0
8.3
9.4
10.2
10.6
0.64
0.86
0.93
0.97
0.98
12.0
14.5
18.0
23.0
33.0
45.0
0.77
0.83
0.88
0.92
0.97
1.00
0.70
0.84
0.90
0.94
0.98
were greater for plot L7 than H4. A general rise in soil salinity in plot
L7, which ended about 3 months before sampling, was followed by a drop in
salinity, particularly at the 0.3- to 0.9-m soil depth, until about 1 month
before sampling. Salinity transients in plot H4 were very small. Thus,
the lower accumulated relative water loss for plot L7 in the 0.6- to 1.2-m
depth may reflect the drop in salinity at that depth before sampling. Re-
gardless, the relative water loss data indicate that two-thirds of the
water is lost above a soil depth of 0.3 m and that 90% is lost above a
depth of 0.9 m. This water loss distribution was used to retarget soil sa-
linity values for irrigation feedback (see Fig. 12) from the salinity
sensors and to determine the intermediate response of soil salinity to the
different leaching treatments.
Response of Salinity Distribution to Changes in Leaching Fraction
Assuming one-dimensional vertical flow, the time-averaged velocity of a
parcel of water within the root zone is given by
v = 6v/9 - {I - E - T /z3dz}/9
o
(3)
where v is the velocity of the parcel of water, 9 is the volumetric water
content (i.e., 6v is the volumetric flux), I is the irrigation rate, E is
the rate of evaporation from the soil surface, T is the rate of transpira-
tion for the volumetric rate of uptake by plant roots. The product of T and
the integral of (3 from 0 to z represents the cumulative rate of uptake above
depth z. The progress of a parcel of water along its path can be calculated
by introducing the expression for v given by equation (3) into
t - t =
o
/z v'1 dz
(4)
-35-
-------
where to is the time at which the parcel was introduced at the soil sur-
face. Use of equations (3) and (4) requires knowledge of I, E, T, (3 and 9.
If I, E, and T are given, the leaching fraction L = (I - E - T)/I is of
course given by implication. Figure 16 shows calculated travel times assum-
ing 1=7 mm/day, 9 = 0.15, and L = 0.05, 0.1, and 0.2. Based on distri-
butions of chloride, the evapotranspiration was assumed to be distributed as
follows:
Soil Depth, z ET Soil Depth, z ET
m % m %
0 to 0.3 70 0.9 to 1.2 4
0.3 to 0.6 10 1.2 to 1.5 2
0.6 to 0.9 8 1.5 to 1.8 1
The sum of the percentages just given is 95%. The remaining 5% represents
an allowance for the fact that the leaching fractions are averages for an
entire tree, whereas the chloride profiles are measured under the tree
canopy where the leaching fraction is relatively large. The time for a
parcel of water to reach 0.9 m ranges from 1 to 2 months and to reach 1.8 m
from 3 to 7 months. In an earlier theoretical study, the distribution of
the uptake was assumed to be given by
3 = S'1 exp (-z/6), (5)
where 6 can be interpreted as a characteristic length for the rooting
depth. In Figure 16, depth-time trajectories for 9 = 0.15, E = 0, L = 0.05,
0.1, and 0.2, and 6 = 0.2 and 0.4 m are shown. The results for 6 = 0.4 m
agree quite well with the travel times based on the uptake distribution
determined from the chloride profiles.
The travel times indicated above must be regarded as minimum estimates.
Farther away from the trunk, the leaching fraction will be smaller. Measur-
ed salinity distributions indicate a continual buildup during the second
year of the experiment.
Vacuum Extractor Volumes and Concentrations
Volume of Leachates
The vacuum extractors were installed in December 1973 in soil that was
relatively wet from previous flood irrigations. Accordingly, some extract
was obtained initially. With the conversion to the new irrigation regime,
and the withholding of irrigation water to increase soil salinity, the roots
dried out the soil at the depth of the extractors and this stored water was
not immediately replenished with water from the surface. As a result, the
tensiometers near the extractors went off-scale and drainage stopped. In
the late spring of 1974, the soil water content increased, the soil matric
potential measured by the tensiometers came back on scale and many of the ex-
tractors started to flow. The 10 and 20% leaching treatments started to
drain before the 5% treatment, as expected. Most of the problems with leaks
in the ceramic tubes were solved. Of the total 36 extractor segments, three
-36-
-------
Travel Time , Days
0 40 80 120 160 200 240 280 320 360
8 =0.4 m
8 * 0.2 m
Figure 16. Chloride-derived and calculated travel time for a parcel of water to pass through the soil
profile as a function of leaching fraction and rooting depth.
-------
are nonfunctional, and two in plot M8 and seven in plot L7 drain very
infrequently or not at all.
Table 8 gives a summary of the cumulative drainage volumes and leaching
percentages calculated from the beginning of the experiment in January 1974,
and the EC of the leachates for the best segment and the average of all
working segments for each of the three treatments. The cumulative leaching
percentages of the best segments as of August 4, 1976 were 4.5, 5.9, and
3.6%, compared to the target values of 5, 10, and 20%, respectively. The
average values were much lower at 1.3, 2.2, and 1.7%, respectively. The
three extractor segments under the southwest tree of L7 had yielded almost
the same drainage volume as the maximum yielding segments of each of plot H6
and plot M8, and much more than those under the remaining three trees of
these treatments. All tensiometers near the extractors read essentially the
same. Therefore, most, if not all, extractors are not functioning properly,
rather than being in dry soil. Since it is unlikely that the extractors
would drain too much, one could take the highest yielding segment as norma-
tive for each treatment. The leaching percentages shown in Table 8 are
lower than those obtained more recently, because of the influence of the
initial period when the extractors were not draining. The leaching percent-
ages from August 1974 to August 1975, and from August 1975 to August 1976,
of the highest yielding segment are 3.9 and 6.5%, 8.1 and 6.9%, and 4.3 and
2.9%, for the 5, 10, and 20% leaching treatments, respectively. The 2-yr
average for the 5% treatment is on target, that for the 10% treatment is 25%
below target, and that for the 20% treatment is 82% below the target value.
These values are not corrected for the fact that the leaching percentages
are based on irrigation volumes as expressed in depth over the total surface
area. Actually, water layers 2.27 times as deep were applied over approxi-
mately 44% of the total soil surface. Therefore, should no appreciable
divergence of flow occur above the depth of the extractors (1.2 m) into the
dry soil between trees, the leaching percentages would actually be only 44%
of the values noted above. On the other hand, should the flow diverge out
to the centerline between trees such that the downward flow at the extractor
depth had become uniform, the leaching percentages would be correct as pre-
sented. The tensiometers in plot H4 indicate that at the depth of the extrac-
tors the flow is still generally downward over an area extending somewhat
beyond the radius of the extractors, whereas flow is outward or even upward
at the extractor depth in a center strip between the tree rows. The down-
ward flow region is largest during the wetter summer season. The extractor
tensiometers on plot H6 measured nearly the same total head as those on plot
H4 at the same depth. This suggests that the flow fields around the extrac-
tors are similar to those measured in plot H4. Thus, there may be little
divergence of flow above the depth of the extractors and the leaching per-
centages cited above probably should be reduced to about half their values.
The tensiometers installed with the extractors to monitor and regulate
the suction in the filter candles of the extractors are very insensitive to
the suction in the filter candles, and read generally the same within the ex-
pected experimental error. Thus, no efforts have been made to adjust the
suction in the extractors to balance the readings between those tensiometers
directly over the extractor and those in the adjacent undisturbed soil.
-38-
-------
TABLE 8. CUMULATIVE DRAINAGE AND LEACHING PERCENTAGE SINCE JANUARY 1974
AND EC OF LEACHATES FOR THE VACUUM EXTRACTORS IN THE THREE LEACHING
TREATMENTS IN THE CITRUS EXPERIMENT
Date
Best Segment
Cumulative
leachate
Cumulative
leaching
EC
Average of all Draining Segments
Cumulative Cumulative Weighted
leachate leaching av. EC
leachate
04-01-74
07-02-74
09-27-74
11-11-74
02-05-75
05-21-75
08-08-75
10-06-75
03-12-76
04-24-76
06-28-76
08-04-76
04-01-74
07-02-74
09-27-74
11-11-74
02-05-75
05-21-75
08-08-75
01-15-76
04-24-76
06-28-76
08-04-76
04-01-74
07-02-74
09-27-74
11-11-74
02-05-75
05-21-75
08-08-75
12-16-75
04-24-76
06-28-76
08-04-76
1.9
7.9
22.5
35.2
39.8
52.5
59.2
70.2
105.5
115.4
131.2
155.9
0.5
1.7
19.4
30.4
38.8
59.3
128.1
166.3
173.0
186.5
246.5
7.2
7.3
60.7
68.5
80.2
97.9
108.8
128.8
143.8
154.1
171.3
S/m n
5% Leaching
S/m
1.1
1.5
2.1
2.9
3.1
3.4
3.0
3.0
4.1
4.3
4.2
4.5
0.5
0.3
1.6
2.3
2.7
3.3
5.2
5.5
5.4
5.0
5.9
7.3
1.2
4 3
*T ~J
4.5
4.8
4.8
4.2
3.9
4.0
3.7
3.6
0.32
0.50
0.41
0.39
0.38
0.44
10% Leaching
0.35
0.27
0.46
0.55
0.41
0.32
0.38
.
20% Leaching
0.27
0.32
0.26
0.29
0.27
0.25
0.2
1.2
4.3
7.6
9.1
11.8
13.8
16.9
26.4
31.2
37.6
46.7
0.2
1.7
12.3
15.8
18.1
24.0
44.3
59.3
62.5
69.1
89.3
1.7
2.5
21.1
24.7
30.9
38.1
44.5
57.1
64.5
70.9
80.2
0.1
0.2
0.4
0.6
0.7
0.8
0.7
0.7
1.0
1.2
1.2
1.3
0.2
0.3
1.0
1.2
1.3
1.3
1.8
2.0
1.9
1.9
2.2
1.7
0.4
1.5
1.6
1.8
1.9
1.7
1.7
1.8
1.7
1.7
0.28
0.52
0.37
0.61
0.61
0.71
__
0.26
0.38
0.73
0.58
0.52
0.65
0.71
0.62
__
""~
0.27
0.31
0.48
0.41
0.37
0.36
0.35
0.47
""~
~~
-39-
-------
Composition of Leachates
Complete salinity analyses of the drainage waters collected by the ex-
tractors are being carried out to ascertain the amounts and compositions of
salt loss by deep percolation resulting from our irrigation management.
To date, 213 samples have been analyzed. Average drainage water composi-
tions by treatment and standard error of mean values for June 1976 samples
are given in Table 9. Compositions with time are compared in Table 10.
These data show that the salt levels in the drainage waters have in-
creased with time and reflect the differences in leaching treatments. The
average total salinities by treatment in June 1976 were (in meq/liter):
113 (plot L7), 75 (plot M8), and 49 (plot H6). The chloride concentrations
were (in meq/liter): 56 (plot L7), 24 (plot M8), and 13 (plot H6), which
corresponded to apparent leaching fractions, Cl. /Cl , of 0.06, 0.15,
and 0.27, respectively. Analogous data in 1974 resulted in leaching frac-
tion estimates of greater than 0.33 for past management. These apparent
leaching fractions represent the degree to which the water percolating past
the 1.2-m soil depth below the irrigated area of the trees has been concen-
trated by evapotranspiration. Since some soil water is currently also
flowing into the inter-row soil regions, the overall LF's may be less than
those suggested by these apparent values. To determine if the shallow
drainage waters sampled are truly representative of the waters that pass
into the ground water, we need to collect and analyze soil-water samples
from deeper depths, both under the trees and in the inter-row regions.
The compositions of the drainage waters in June 1976 had not equaled
those expected for steady-state conditions. The expected composition of the
drainage water as a function of leaching fraction is given in Table 11. Com-
pared to these compositions, the waters in June were high in calcium and low
in magnesium, sodium, sulfate, and SAR for all treatments. This probably
resulted from the fact that sodium and magnesium were still being adsorbed
by the exchange complex and calcium was being displaced to solution. Ex-
change equilibrium had not been achieved. This is reasonable, considering
the short time the experiment has been underway.
Yield and Fruit Quality
The average yields per tree for the three leaching treatments are
presented in Table 12, along with the average yield for the border and
check-plot trees. The average yield per tree for each replication of the
experiment and for each row of the border and check-plot trees is given in
Appendix Table A-2. Analysis of variance showed no significant differences
in yield among leaching treatments for each of the 3 yrs. This lack of sig-
nificant yield differences among leaching treatments, however, was not
unexpected, since significant yield differences did not appear in an irriga-
tion water quality trial on orange trees until the fourth year in a study
by Bingham et al. (1974). Yields from the borders and check plots were not
significantly different from the experimental trees and they were more vari-
able. Yields in 1975 and 1976 were consistently larger from the fertilizer-
check plot and the drip border than from the leaching treatments. Histori-
cally, the east side of the grove, where the drip border and fertilizer
check plot are located, has out-yielded the west side.
-40-
-------
TABLE 9. AVERAGE DRAINAGE WATER COMPOSITIONS FROM THE VACUUM EXTRACTORS
FOR THE CITRUS EXPERIMENT IN JUNE 1976
Plot
no.
L7
M8
H6
Leaching
treat-
ment
5%
10%
20%
EC*
S/m
0.86 f
(0.09)
0.58
(0.06)
0.39
(0.02)
Ca
54.9
(9.5)
27.2
(3.0)
21.0
(2.0)
*
Electrical conductivity at
t . ,- _._.,
Mg
V
22.0
(3.1)
14.2
(1.9)
8.8
(0.6)
25°C.
Na K
35.5 0.39
(4.1) (.03)
33.2 0.52
<3.7) (.04)
18.9 0.29
(1.3) (.02)
Sum of
cations
- meq/£ -
112.8
(11-6)
75.2
(7.8)
48.9
(2.6)
HC03
v
6.2
(0.6)
8.2
(0.6)
8.6
(0.7)
S°4
48.5
(4.9)
42.8
(5.3)
27.0
(1.9)
Cl
55.6
(13.4)
24.3
(3.5)
13.1
(1.3)
N03
2.6
(0.8)
1.1
(0.2)
0.7
(0.1)
Si02
mg/A
35.2
(2.5)
33.4
(2.2)
35.8
(3 ..2)
SARt Cl. §
iw
Cl,
dw
6.3 .06
(0.9)
7.4 .15
(0.6)
5.1 .27
(0.5)
SAR = Na//(Ca + Mg)/2, where solutes are in meq/liter.
o
Ratio of Cl concentration (meq/liter) in irrigation water (iw) to drainage water (dw).
Standard error of mean.
-------
ho
TABLE 10. COMPOSITION OF DRAINAGE WATERS FROM SELECTED VACUUM EXTRACTORS
IN THE CITRUS EXPERIMENT AS A FUNCTION OF TIME
Leaching
treatment
7
5(L7)
10 (M8)
20(H6)
Extrac- Date
tor No.
B-l 07-73
02-75
06-76
B-l 04-74
06-75
06-76
B-l 08-74
08-75
06-76
Ca
9.2
17.2
79.3
9.7
21.0
37.4
22.2
34.2
29.6
Mg
5.5
9.8
27.6
6.1
12.4
21.1
11.0
15.0
10.7
Na
10.3
15.9
23.9
10.4
26.1
29.5
16.0
14.3
20.5
K
0.2
0.3
0.3
0.3
0.4
0.6
0.3
0.2
0.3
Sum of
cations
25.1
43.3
131.0
26.5
60.0
88.7
49.4
63.8
61.1
HC°3
7.5
7.4
4.9
6.8
10.5
8.9
5.8
6.4
7.1
SO.
4
9.4
18.7
44.1
11.0
35.2
46.4
24.0
33.7
35.1
Cl
7.6
17.3
82.0
9.2
12.7
31.2
17.1
22.8
18.6
N03
0.6
1.3
3.2
0.3
1.2
2.0
1.4
0.7
0.8
Si02
mo/P
mg/x,
95
73
24
77
22
25
81
61
38
EC*
C /-m
0.23
0.37
1.01
0.24
0.45
0.66
0.41
0.47
0.47
SARt
3.8
4.3
3.3
3.7
6.4
5.5
3.9
2.9
4.6
Electrical conductivity at 25°C.
SAR = Na//(Ca - Mg)/2, where solutes are in meq/liter.
-------
TABLE 11. PREDICTED DRAINAGE WATER COMPOSITIONS FOR CITRUS TREATMENTS*
Leaching
fraction
.03
.05
.10
.15
.20
.30
.40
Ca Mg
23.7 90.0
24.9 54.0
22.8 27.0
17.2 18.0
14.4 13.5
11.6 9.0
10.1 6.7
*
Assuming the partial pressure
is Ca (4.5), Mg (2.7), Na (6.
Electrical conductivity at 25
8 ,, - .
Na
227.0
136.0
68.0
45.3
34.0
22.7
17.0
of CO-
8), cr
°C.
Sum of
cations
_____ man / 0 ____
340.0
215.0
118.0
80.6
61.9
43.2
33.9
is 0.015 and the
(3.5), HC03 (2.8)
HC03
8.5
6.9
5.8
5.9
5.9
5.9
5.9
so4
215.0
138.0
77.0
51.3
38.5
25.7
19.2
irrigation water
, and SO, (7.7).
Cl
117.0
70.0
35.0
23.3
17.5
11.7
8.7
composition
ECt
S/m
2.46
1.68
1.00
0.69
0.53
0.37
0.29
SAR§
30.1
21.7
13.6
10.8
9.1
7.1
5.9
(in meq/liter)
SAR = Na//(Ca + Mg)/2, where solutes are in meq/liter.
-------
TABLE 12. AVERAGE ANNUAL VALENCIA ORANGE YIELD (kg/tree) FOR THE MINIMUM
LEACHING TREATMENTS AND THE BORDER- AND CHECK-PLOT TREES
Year Leaching treatment Border trees Check plots Weighted
1974
1975
1976
5%
116
143
74
10%
126
148
70
20%
122
147
71
Bubbler
123
141
77
Drip
113
165
88
Flood
*
121
98
Pert.
A
179
105
average
120
149
83
Trees were not individually harvested.
The yields reported here and the historical yield record of the grove
indicate an alternate year pattern in fruit bearing. This yield pattern
for Valencia orange is well documented (Parker and Batchelder, 1932;
Bingham et al., 1974). Thus, although the yields of 1976 were discourag-
ingly low, they were not unexpected.
The average number of fruit harvested per tree is summarized in Table
13. The number of fruit harvested each year agrees well with the weight of
fruit harvested, but the variation among years for the number of fruit is
even larger than variation among weights. The numbers of fruit per kg each
year were 4.5, 5.2, and 4.2 for 1974, 1975, and 1976, respectively,
indicating smaller fruit for the higher yields.
TABLE 13. AVERAGE NUMBER OF VALENCIA ORANGE FRUIT HARVESTED PER TREE FOR
THE MINIMUM LEACHING TREATMENTS AND THE BORDER- AND CHECK-PLOT TREES
Year
1974
1975
1976
Leaching treatment
5%
522
788
329
10%
558
798
299
20%
559
781
309
Border
Bubbler
551
706
314
trees
Drip
479
871
349
Check
Flood
519
426
plots
Pert.
__
939
406
Weighted
average
537
770
349
The average annual value of each measure of fruit quality is presented
in Table 14. None of the fruit quality measurements differed significantly
among the leaching treatments in any year. The fruit-size data support the
results from the number of fruit per kg in that the heavier yield in 1975
resulted from smaller fruit (larger ring size, shorter, and narrower than
in 1974 and 1976). There were few differences among the leaching treat-
ments, borders, and check-plots. In 1974, the fruit from the drip border
were consistently larger than the fruit from the other treatments. Fruit
from the bubbler-border and the flood-check plots were larger than the
fruit from the remaining treatments in 1975. No consistent differences were
noted in rind color. The orange rind was consistently rougher in 1976 than
in either 1974 or 1975 for all treatments, and in 1975, the fruit from the
drip border were smoother and the fruit from the flood-check plot were
rougher than from the remaining treatments.
-44-
-------
TABLE 14. VALENCIA ORANGE FRUIT QUALITY FOR 1974, 1975, AND 1976 HARVESTS
Ul
I
Measure of
fruit quality
*
Ring Size
t
Rind Color
8
Rind Texture
,
Fruit
Length, mm
Fruit
Width, mm
Year
1974
1975
1976
1974
1975
1976
1974
1975
1976
1974
1975
1976
1974
1975
1976
Leaching treatments
5%
70
82
70
10.3
10.4
10.4
3.3
3.2
4.1
80
77
84
74
72
76
10%
68
81
68
10.3
10.4
10.3
3.3
3.3
4.2
81
78
85
75
72
. 77
20%
69
79
71
10.3
10.4
10.3
3.3
3.3
4.2
81
78
84
74
73
76
Border trees
Bubbler
72
75
65
10.2
10.4
10.5
3.2
3.4
4.3
81
81
88
74
75
77
Drip
64
86
63
10.1
10.3
10.3
3.5
2.9
4.2
84
78
86
77
71
79
Check plots
Flood
75
68
70
10.4
10.4
10.4
3.3
3.8
4.2
78
84
84
72
77
76
Fertilizer
81
66
10.4
10.3
__
3.3
4.3
__
78
85
__
72
76
Weighted
Average
69
79
68
10.2
10.4
10.4
3.3
3.3
4.2
81
79
85
74
73
77
Ring size is equivalent to the average number of fruit to a standard size shipping carton.
Rind color is denoted by a numerical scale from 3 (dark green) to 13 (orange-red).
p
Rind texture is evaluated on a numerical scale from 1 (smooth) to 6 (very rough) .
-------
Results of the annual juice analyses are given in Table 15. The fruit
were harvested purposely at a lower TDS/TA ratio in 1975 and 1976 than in
1974. As orange fruit ripens, the TDS/TA ratio increases because TDS
decreases slower than does TA. The legal minimum ratio for sale of oranges
is eight. Juice quality measurements did not differ significantly among
leaching treatments in any year. Fruit were not as ripe from the flood-
check plot as from the other treatments, as is evident from a lower TDS/TA
ratio and a lower percent juice.
Leaf Analyses
Leaf samples, taken in September 1974 and 1975 for evaluating the
nutritional status of the trees, were analyzed for 13 mineral elements. The
1975 analytical results for the experimental, border, and check plots are
summarized in Table 16. The 1974 results are given in Table A-3. The
nutritional status of the trees was reasonably uniform among all plots and
all nutrient elements except boron were within or near optimum levels. Leaf
boron concentrations were unexpectedly high and will require careful monitor-
ing in the future. Analyses of the soil and irrigation water did not indi-
cate either as the boron source. Sodium and chloride, which can accumulate
to toxic levels, were well within acceptable limits. However, Cl levels
seem to be increasing in the low leaching treatments.
Except for some chlorosis observed in the spring on new growth and at-
tributed to seasonal deficiencies in Fe and Mn, the trees appeared in good
health. Leaf samples were taken again in October 1976 for continued evalua-
tion of tree nutrition and the fertilizer program.
Trunk Circumference
Trunk circumferences were measured in September 1973, July 1974, August
1975, and July 1976. In 1974, a nail was driven into each trunk to locate
permanently the measurement elevation. The average annual increases in
trunk circumference for the leaching treatments, the borders and the check
plots are given in Table 17. The annual measurements of circumference,
averaged by replication for the leaching treatments and by row for the
borders and check plots , are given in Table A-4.
There have been no significant differences in trunk circumferences among
leaching treatments. Analysis of variance showed, however, that the mean
trunk circumferences for all three leaching treatments have enlarged signif-
icantly each year with the exception of the 5% leaching treatment during the
initial year. The mean trunk circumference of the experimental trees in-
creased less than 2.5% the first year, but more than 5% during each of the
last 2 yrs. Although less consistent than the experimental trees, the
border trees have made similar growth. Measurements were not taken every
year on the check-plot trees, so comparisons cannot be made. However, they
seem to be growing as rapidly as the experimental trees, although they are
smaller and more variable in size.
-46-
-------
TABLE 15.
Measure of Year
juice quality
Total dissolv-^
ed solids, %
Total .
acid, %
Ratio ,
TDS/TA
Percent
j uice
Extractable
TDS, %§
1974
1975
1976
1974
1975
1976
1974
1975
1976
1974
1975
1976
1974
1975
1976
VALENCIA CHANGE JUICE QUALITY FOR 1974, 1975,
Leaching treatments
5%
9.7
10.3
9.9
0.90
1.03
0.99
10.8
10.0
10.0
44.3
43.7
41.4
3.65
3.83
3.45
10%
9.8
10.1
9.9
0.89
1.01
0.99
11.0
10.0
10.0
44.1
M.2
40.9
3.67
3.71
3.40
20%
9.8
10.0
9.8
0.90
1.00
0.98
10.9
10.0
10.0
44.6
42.7
40.6
3.72
3.63
3.40
Border trees
Bubbler
10.0
10.0
9.8
0.89
0.97
0.94
11.2
10.3
10.5
44.3
43.1
38.4
3.77
3.66
3.10
Drip
9.1
10.1
9.7
0.83
1.00
0.93
11.0
10.1
10.3
42.5
43.9
41.8
3.29
3.77
3.45
AND 1976
HARVESTS
Check plots
Flood
10. 0
10.0
10.2
0.93
1.04
1.10
10.7
9.6
9.2
41.8
39.3
36.6
3.55
3.34
3.20
Fertilizer
10.1
9.7
1.02
0.97
9.9
10.0
43.3
41.6
3.72
3.45
Weighted
average
9.8
10.1
9.9
0.89
1.01
0.99
10.9
10.0
10.0
43.6
42.9
40.2
3.62
3.66
3.36
Total dissolved solids (TDS) are determined with a Brix hydrometer or refractometer. Total dissolv-
ed solids in orange juice consist mainly of sucrose, fructose,. and citric acid, together with potas-
sium, calcium, magnesium, and sodium salts. Traces of glycosides are also present.
Total acid refers to the titratable acidity and is determined volumetrically by the amount of
standard alkali necessary to neutralize the acid in a known amount of juice. The acid in orange
juice consists mainly of citric acid with small amounts of malic, tartaric, and succinic acids.
The juice plant can extract only about 85% as much juice as can the laboratory. Thus, the
grower is paid on the basis of 85% of the total dissolved solids per fruit weight measured in the
laboratory. To convert from % (given above) to Ibs. of solids per ton of fruit, multiply by 20.
-------
TABLE 16. MINERAL COMPOSITION OF VALENCIA ORANGE LEAVES SAMPLED SEPTEMBER 1975
00
Element (unit) Optimum
range
N
P
K
Ca
Mg
Na
Cl
S
B
Fe
Mn
Zn
Cu
% 2.2-2.7
mmole/100 g 4-6
meq/100 g 25-45
" " 150-250
" 17-50
2.5-6.5
<4.2
12-19
Ug/g 25-150
" 60-150
" 20-100
" 25-100
" 4-10
Leaching treatments
5%
2.9
4.1
31
224
23
1.6
3.4
20
208
56
26
28
9
10%
2.8
4.1
30
224
23
1.7
3.0
20
212
58
26
27
9
20%
2.8
4.2
32
221
22
1.6
1.8
20
206
58
28
27
9
Border
Bubbler
2.7
4.2
32
234
23
1.6
1.9
20
182
56
31
28
10
trees
Drip
2.8
4.4
31
227
21
1.8
2.2
20
182
62
25
27
9
Check plots
Flood
2.5
3.6
34
236
22
0.6
0.6
21
195
61
22
17
10
Fertilizer
3.1
3.9
31
210
19
1.7
1.2
20
211
65
25
21
8
-------
TABLE 17. AVERAGE ANNUAL INCREASE IN THE TRUNK CIRCUMFERENCE (mm)
OF VALENCIA ORANGE TREES FOR THE MINIMUM LEACHING
EXPERIMENT IN SOUTHWESTERN ARIZONA
Growth
period
1973-74
1974-75
1975-76
Leaching treatment
5%
1Qnst
34**
36**
10%
*
15
**
33
**
37
20%
16*
**
35
**
38
Border trees
Bubbler
13
40
25
Drip
22
28
38
Check plots
Flood Fertilizer
__. __.
43
22 24
Results of the analysis of variance comparing the nine replication means;
* **
and denote statistical significance at the 5 and 1% levels, respec-
tively.
Soil Air
Soil Oxygen
The soil oxygen results are summarized in Table 18. Soil oxygen con-
centrations above 10% are considered adequate for most crops. No measure-
ments were below 10% in the citrus experiment, not even in the flood check
1 day after an irrigation. Thus, poor soil aeration is not expected to be
a problem with trickle irrigation of citrus on these coarse-textured soils.
TABLE 18. COMPARISON OF THE PERCENT SOIL OXYGEN AT THE 0.45-m SOIL DEPTH
AND 1.5 m OUT FROM THE VALENCIA ORANGE TREE TRUNK
FOR SEVERAL IRRIGATION TREATMENTS
Date Leaching Treatments Border Trees Flood Check
5% 20% Bubbler Drip
Feb. 1975 17.4 20.1 19.8 20.4 20.2
June &
July 1975 18.1 14.8 18.2 18.5 20.2
Soil Carbon Dioxide
The carbon dioxide concentrations in the soil air (Table 19) are af-
fected by time of year, but not by soil depth. They are higher during the
summer months (May-October) than in January because of increased root and
microbial respiration and increased soil-water content. The former results
in greater rates of carbon dioxide production and the latter increases the
resistance to carbon dioxide exchange with the atmosphere.
-49-
-------
TABLE 19. AVERAGE PERCENT CARBON DIOXIDE IN SOIL AIR UNDER
MATURE CITRUS TREES AS A FUNCTION OF TIME AND SOIL DEPTH
Soil Depth
(m)
0.3
0.6
0.9
1.2
Oct 1975
2.8
3.0
3.0
3.0
Jan 1976
0.4
0.4
0.6
0.6
Mar 1976
1.7
1.5
1.7
1.5
May 1976
2.5
3.2
2.4
2.6
DISCUSSION
Our experience thus far suggests, but does not establish with cer-
tainty, that citrus can be irrigated with leaching percentages of 20 or less
without decreasing production. Available data indicate that dynamic
equilibrium was approached in 1975, so that evapotranspiration can be rea-
sonably estimated. Unless the trees respond physiologically after a longer
time - a possibility that cannot be overlooked - the data also permit esti-
mates of the effect of changing irrigation management on drainage volumes
and salt loading in the drainage waters.
Evapot ransp ira t ion
Based on the results presented, evapotranspiration (ET) for citrus can
be estimated by several techniques. First, ET can be calculated by sub-
tracting the amount of leaching from the depth of irrigation and rainfall
(I). This assumes the desired leaching is achieved, an assumption that is
reasonable in view of the carefully controlled differentials and the inter-
nal consistency of the data. A second approach is to calculate the leach-
ing percentage from the ratio of the chloride concentration in the irriga-
tion water (Cliw) to that at the bottom of the root zone (Cl, ) as
measured from soil samples or vacuum extractor leachates for individual
plots; ET is then determined as ET = I[l ~(Cliw/Cldw)]. The response
time of soil salinity to decreased leaching was slower than anticipated,
but since 1975 steady state has been achieved basically so that chloride
concentrations at the bottom of the root zone should be representative of
the leaching treatments. A third technique is to calculate ET from meteor-
ological data. A fourth method would be to use the drainage volume as :
measured in the vacuum extractors. However, this last method will not be
used, because the volumes collected have been small and inconsistent.3
Here we compare the three estimates of ET with each other and with published
values.
3We believe that the ceilings of the extractor tunnels were smeared
during formation. Some extractors will be removed for examination early
in 1977.
-50-
-------
TABLE 20. ESTIMATE OF CITRUS EVAPOTRANSPIRATION (ET) BASED ON CHLORIDE CON-
CENTRATION FROM VACUUM EXTRACTORS OR SOIL SAMPLING
Date C1 Source Water applied Leaching Annual
sampled -c '
H4
L7
H6
M8
L7
Mar 1976 soil sample
ii ii ii M
Jun 1976 vacuum extractor
ii ii n ii
ii it n * n
mm
1730
1340
2066
1706
1456
ci. /ci.
iw dw
0.30
0.06
0.22
0.12
0.05
Average
E,±
mm
1211
1260
1611
1501
1383
1390
The estimate of ET based on water application and the desired leaching
during the past 3 yrs is 1434 mm. The estimate of ET determined from those
plots where chloride concentrations were measured is 1390 mm (see Table 20).
These data are based on chloride concentrations from soil samples taken in
March 1976 for plots H4 and L7 and upon chloride concentrations measured in
extractor leachates in June 1976 for plots H6, M8, and L7. The leaching per-
centages based upon leachate composition were reduced 17% to correct for the
effect of vacuum extractor location with respect to average chloride concen-
trations found from soil samples. The modified Penman and the Jensen-Haise
equations, based on meteorological data, led to annual ET estimates of 1421
and 1342 mm, respectively.
The three estimates of ET closely agree and substantiate an annual
value of 1400 mm. The resulting leaching percentages for the three leach-
ing treatments from January 1974 to September 1976, based on this ET and
the amounts of water applied, are 8, 11, and 22. Likewise, the leaching
percentage for the flood check is 47.
Erie et al. (1965) published an estimate of 1000 mm for Navel oranges
at Phoenix, Arizona; this value is significantly lower than ours. A likely
explanation for the difference is that water use from the top 150 mm of soil
was not taken into account in the evaluation reported by Erie et al. (Jensen,
M. E., personal communication, 1975).
Salt Load Reduction
To estimate the effect of changing irrigation management of citrus on
the contribution to the drainage volume and its salt load from the Wellton-
Mohawk project, one may make the following assumptions: (1) annual ET is
1400 mm; (2) salt concentration of the irrigation water (S ) is 944 mg/Jt
(see footnote #2, p. 8); (3) present water delivery to the ^000 ha of citrus
(A) is 3200 mm/yr (I ); and (4) the salt concentration of the groundwater
pumped at present avlrages 3000 ag/Z (S ). We will further assume no
decrease in irrigation application effi§₯ency below that'of the experiment.
-51-
-------
The annual reduction in the volume of drainage water resulting from a
LF of 0.2 would be
AV = A{l - [ET/U - LF)]}
= 3000 [3200 - (1400/0.80)]
= 4.35 x io7 m3/yr
= 35,000 acre-feet/yr.
Assuming no effective salt precipitation, the reduction in salt load of the
drainage water leaving the root zone would be
A salt load = S. AV
iw
= 41,000 Mg/yr
= 45,000 tons/yr.
Accounting for lime precipitation (Oster and Rhoades, 1975), the salt load
would be reduced an additional 11%, or
A salt load = 45,500 Mg/yr.
The actual amount of salt exported in the Wellton-Mohawk drains, pre-
sumably the feed water to the proposed desalting complex, would depend on
changes in groundwater composition with time. During the next several
years, the groundwater quality should not change significantly. Thus, the
effect of changing citrus irrigation management on the quantity of salt ex-
ported would be a reduction of 3000 mg/£. x (4.35 x 1Q7 m3/yr) or 130,000
Mg/yr. These estimates may be compared with a current annual drainage
volume of approximately 26 x IO7 m3/yr containing 780,000 Mg/yr of salt.
-52-
-------
SECTION 4
ALFALFA
EXPERIMENTAL PROCEDURE
Experimental Design
The alfalfa experiment is located 13 km northeast of Tacna, Arizona, on
the Snyder Ranch in the flood plain of the Gila River. The experimental
site is the northern quarter of an 8-ha field that had been previously
cropped to alfalfa and flood irrigated. Within 400 m of the south end of
the field, the topography rises more than 15 m to a mesa where citrus is
grown. A drainage well for water table control is adjacent to the southwest
corner of the field. The experimental site was instrumented and the irriga-
tion system installed during September 1974. After cultivation, alfalfa
(Medicago sativa L. , cv- Hayden) was planted in early October. All crop
management practices, such as cultivating, planting, harvesting, and pest
and*weed control, are performed by Snyder Ranch personnel.
The experimental design, shown in Fig. 17, consists of 5, 10, and 20%
leaching treatments, each replicated five times. Each plot is 12.2 m wide
and 104 m long. The remainder of the 8-ha field is irrigated by the rancher
using level-basin flooding.
Soil Properties
Soil Conservation Service personnel examined several soil profiles
within the experimental site and classified the soil as Indio fine sandy
loam (Typic Torrifluvent, coarse-siIty, mixed, hyperthermic). The soil de-
scription prepared by the SCS for one of the soil profiles is given in
TABLE B-l of the appendix. The soil is calcareous, moderately alkaline,
well drained, and representative of the valley soils in the district. The
soil profile texture grades from very fine sandy loam to silt loam at a
depth of 0.3 m, and to silty clay loam at a depth of 0.8 m.
Following the first harvest in February 1975, 303 soil samples were col-
lected to characterize initial soil conditions and determine the soil
salinity distribution resulting from previous irrigation management. Six
locations were sampled in each of the six instrumented plots. These 36
locations were sampled by 0.3-m increments, 21 to a depth of 1.5 m and 15 to
a depth of 2.4 m. Twelve locations were sampled in the level-basin flood-
irrigated check, six to a depth of 1.5 m and six to 2.4 m. All of the
samples were subjected to the same analyses as those collected in the citrus
-53-
-------
LEGEND
Plot borders and location of wheel tracks of irrigation system
M10 Replication number and leaching percentage
* Instrumentation Manhole 17m deep X 1.5m dia.
75mm I.D. Conduit
Ul
c
jo
s2
in
c ,
1 1 1
I 1
1
1 1
rl
JI20
a>,
TJI
ol
00,
"1
1
1
1
i
1
1
1
1
1
15
1 1
i i
no
H20
n5
D
Jl2.l9m|«-
nio
v_
1
1
m2qE5
1
in 10
i j
i
1
IV20
IV 5
|
1
1
TVIO
V20
V5
VIO
a>
^
o
00
Figure 17. Design of minimum leaching experiment for alfalfa in southwestern Arizona.
-------
experiment. As in the citrus, the relationship between hydraulic conductiv-
ity and matric potential was determined in the laboratory on three undis-
turbed soil columns, each 0.9 m deep, taken in the fall of 1974.
Irrigation
The irrigation water, pump, and filter system are identical to those used
for the citrus. Water is applied with an electrically driven, lateral-move
irrigation system 195 m long, with wheel towers spaced 12.2 m apart (see Fig.
18). The irrigation system was operated with impact sprinklers spaced at
12.2-m intervals along the main line for the first month after planting. On
Nov. 8, 1974, after the crop was well established, the spray system shown in
Fig. 18 was installed and differential irrigation treatments were initiated.
During the first month of operation, several adjustments in alignment of the
system were required to keep it running in the same tracks. Problems also
arose from water running into the wheel tracks, causing the system to mire
down. By filling the tracks with sand and by shielding spray nozzles near
the wheels, these problems have been overcome. The irrigation system has
operated trouble-free for over a year.
Differential irrigation treatments are achieved by regulating the pres-
sure in each section of submain with a throttle valve. By periodically ad-
justing the pressure, based on water meter readings, the average differences
in accumulated water application between the three treatments have been
within 0.2% of those desired. Uniformity of water distribution within plots
is also good. The standard deviation of water delivery rate from individual
spray nozzles was about 2%, and the standard deviation of water delivery
between 2.4-m-wide sections of a plot (irrigated with four spray nozzles)
was within 1%. By selectively interchanging up to one-sixth of the noz-
zles, the differences in water delivery rate between 2.4-m-wide sections
of any plot were reduced to less than 0.1%. This is well within the re-
quired uniformity and is possibly the most uniform field irrigation system
in existence.
As in the citrus experiment, the water for each plot is measured by two
water meters in series. The water is distributed through 20 nozzles (Fig.
18) , which, at a pressure of 70 kPa, deliver 30 ml/sec each. At a ground
speed of 30 m/hr, the irrigation system applies 6 mm each pass. Thus, the
number of passes varies from one every few days in winter to a maximum of
two per day in summer.
Instrumentation
Two replications of each leaching treatment were instrumented from
three concrete manholes, each installed at the border between two_plots as
shown in Fig. 17. The manholes facilitate automatic data collection and
minimize interference with farming operations. A conduit from the service
building at the edge of the experimental plots to the manholes shown in
Figs. 17 and 19, contains electrifies or automa ic^ata collection,^
^^T2.sr;; si.*: d*
depth in each of the six instrumented plots.
-55-
-------
TOP VIEW
Water
meters
Drip
Valve
6-mm 1.0. PVC Pipe
i i \A\ i
r'lYT') i i
Spray Nozzle
Wheel
Drip
SIDE VIEW
00-mm I.D Aluminum Mainline
18-mm I.D. Polyethylene hose
I.D. PVC Sub
Main
Figure 18. Spray system for one section of the lateral-move irrigation system serving one replication.
-------
Ul
X X
Extractor j"
^-Conduit Trench ' /y
' i Jr~\
1 * jTj
u
x x
L_._j
.Plot Border and Wheel Track
~ of Irrigation Tower
x x
Excavated Pit
^ 1 """" "
H". J ' J "' -_ _ m - n . _[
fl i ^- conduit Tunnel
Manhole] »"
x x
LEGEND
Scale: <"P »
x Salt Sensors O.3 , 0.6. 0.9, Q
Tensiometers 0.3, 0.6, 0.9,1.2
j 1
j j
Suction Lysimeter
1.2 m deep
, Q 1.5 m deep
Figure 19. Location of instruments for two of the six plots
instrumented in the alfalfa experiment.
-------
Salinity sensors and tensiometers were installed as shown in Fig. 19.
All the tensiometers installed in September 1974 had to be replaced because
epoxy cement failed to seal the ceramic tensiometer cups to an acrylic
adapter. New tensiometers were installed in February 1975. All sensor
wires and tensiometer tubes leading to the manholes were buried at least
0.3 m below the soil surface.
Initially, the 40 tensiometers near each manhole (including 16 over the
vacuum extractors) were connected to two switching boxes housing 24-position
hydraulic scanning valves, each controlled by an electric stepping switch.
The pressure within each tensiometer, as well as that of a standard and zero
reference, was measured with a pressure transducer producing a voltage
signal. These voltages were converted to soil hydraulic potentials by the
data acquisition system. A calculator program was written to read the 120
tensiometers and references at any desired time interval. All tensiometers
and associated tubing could be flushed and recharged automatically whenever
it was necessary to eliminate air bubbles. The first automatic tensiometer
data were obtained in March 1975. Measurements were continued through
September, but were often interrupted or inaccurate because of problems with
the scanning valves. After September, these valves became completely unsat-
isfactory and the entire automatic scanning system had to be discarded.
Since March 1976, all tensiometers have been connected to mercury manometers
that are located within the manholes and are read and recorded manually.
The coaxial tubing for each tensiometer and a common shutoff device allow
the tensiometers in a manhole to be flushed simultaneously. The tensiom-
eters appear to be working satisfactorily.
Since January 1975, the salinity sensors have been read with the data
acquisition system. Through a computer program, the unit activates the
stepping switches, reads all 96 sensors in sequence, compensates the read-
ings for soil temperature, and prints out the calculated soil water elec-
trical conductivities on paper tape. Readings taken automatically are nearly
identical to those read manually. It requires 20 minutes to read the sensors
and produce the punched paper tape, which is then transmitted via teletype to
Riverside for processing and analysis.
In addition to the vacuum extractors, a 1.2-m-deep suction lysimeter
was installed in each of the six instrumented plots. These were formed by
lining a 1.5-m-square hole with 250-|Jm-thick plastic film. A wooden frame,
0.3 m below the soil surface, serves as the top lip of the lysimeter. After
backfilling with about 0.1 m of soil, ceramic tubes, identical to those used
in the vacuum extractors, were installed in each lysimeter to extract the
leachate. The lysimeters were then filled with disturbed soil and compacted
by saturating with water.
Check Plot
A portion of the 8-ha field not within the experimental area serves as
a check and is irrigated as a level basin from a single gate located in one
corner of the field. During each irrigation, a depth of about 150 mm is
applied at the rate of about 0.4 m Is. Irrigation volume is measured with a
-58-
-------
critical-flow flume identical to that used in the citrus experiment. Salin-
,ity sensors were installed at four depths (0.3,.0.6. 0.9, and 1.2 m) in six
locations in the check plot.
Yield
The alfalfa is harvested by cutting with a windrower-conditioner and
then cubed following customary procedures. Harvest dates are chosen by the
ranch when the alfalfa is at approximately 50% bloom. The yield of each
cutting is determined by weighing three samples from each replicate. The
samples are taken from 26.8-m2 areas in the north, center, and south sec-
tions of each plot from the first swath (4.4 m wide) made down the center
of each plot. Nine similar samples are taken from the flood check at three
locations in each of three windrows made in the east, center, and west sec-
tions of the field. Each sample is weighed and yields of air-dry alfalfa
are calculated after measuring the water contents of representative sub-
samples.
RESULTS
Soil Properties
Only about three-fourths of the initial soil samples have been anal-
yzed. A preliminary summary of the soil properties for Indio fine sandy
loam is given in TABLE 21. It shows that the soil is nonsaline (EC less
than 0.4 S/m) through most of the sampled profiles, indicating that this
field has been well leached in the past. The chloride concentration in the
lower depths is about 10 meq/liter in the saturation extract. Assuming a
Cl concentration of 3.5 meq/liter in the irrigation water, this corresponds
to a leaching percentage of about 35. Salt concentrations generally peak
at a depth above 0.6 m, suggesting the effective root zone is relatively
shallow or that water penetration is impeded by the textural break.
The relationship between hydraulic conductivity and matric potential
for Indio fine sandy loam is given in Fig. 20. The length of the horizontal
lines indicates the range of matric potential measured for a given hy-
draulic conductivity.
Water Use
The average daily rates of water application, rainfall included, are
given in Fig. 21 for the three leaching treatments and the flood check from
January 1975 to July 1976. Each data point in the summer is the daily aver-
age for an individual cutting, typically covering 4 to 6 weeks. Data points
during the winter are the averages for about 2 months. The total depths of
water applied over the 19 months shown in Fig. 21 were 3152, 3322, and 3613
mm for the 5, 10, and 20% leaching treatments, respectively, and 3598 mm
for the flood check.
As with the citrus, the evapotranspiration rate (ET) can be estimated
by subtracting the planned leaching depth from the actual depth of water
-59-
-------
TABLE 21. REPRESENTATIVE SOIL PROPERTIES FOR THE SNYDER RANCH ALFALFA FIELD
Property Unit
Cation-exchange- meq/lOOg
capacity
Exchangeable-sodium- %
percentage
Sodium-adsorption- -
ratio
Saturation g/lOOg
percentage
Field water g/lOOg
content
PH
EC S/m at 25°C
Soluble* calcium meq/liter
Soluble magnesium "
Soluble sodium "
Soluble potassium "
Soluble bicarbonate "
Soluble sulfate "
Soluble chloride "
Soluble nitrate "
Depth
(meter)
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0 -
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
Typical
values
15
28
8
10
6
7
40
50
14
28
7.4
7.5
0.25
0.35
5
15
3
8
12
20
0.3
0.3
2.3
2.5
16
28
5
10
0.6
0.3
Range
10 -
25 -
5 -
5 -
5 -
5 -
30 -
42 -
11 -
23 -
6.9 -
7.0 -
0.15-
0.17-
3.8 -
2 -
3 -
1 -
10 -
10 -
0.1 -
0.1 -
1.9 -
1.4 -
8 -
15 -
3 -
5 -
0.2 -
0.1 -
19
35
12
30
8
30
45
60
30
36
8.2
7.8
0.5
1.0
26
30
16
18
27
60
0.7
0.6
4.0
6.0
30
80
10
30
3
1.7
Soluble in saturation extract.
-60-
-------
Matric Potential , kPa
30 -20
5000
Figure 20. Relationship between .soil matric potential and hydraulic conduc-
tivity for Indio fine sandy loam.
-61-
-------
. » o
T3
IE
o c
tr =
<0
o
0>
o
o
a.
CL
o>
I i i i i i i i I i r I I i I I
J FMAMJ JASONDJ FMAMJ J
1975 1976
Figure 21. Total water application to the three leaching treatments and
the flood check for the alfalfa experiment.
-------
applied. This uncorrected daily evapotranspiration rate for alfalfa com-
puted monthly, is presented in Fig. 22. For comparison, the average'daily
rate of pan evaporation is also given. This evaporation pan is located in
the valley about 15 km west of the alfalfa experiment. Based on these data
the pan factor (ratio of ET to pan evaporation) for alfalfa is 0.78. The
average annual ET estimated from the data for 19 months is 1813 mm, and
ET based on the year from July 1, 1975 to July 1, 1976 is 1740 mm.
As independent checks, the ET was also computed by the modified Penman
equation (Doorenbos and Pruitt, 1975), and derived from measurements from a
weighing lysimeter in the Imperial Valley of California, and from Erie et al.
(1965) . All the results are given in TABLE 22 as average daily ET rates by
month. The three independent estimates of ET are from 4 to 21% higher than
TABLE 22. COMPARISON OF VARIOUS TECHNIQUES OF ESTIMATING EVAPOTRANSPI-
RATION OF ALFALFA IN SOUTHWESTERN ARIZONA
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Yearly
Erie* Brawleyt
et al. lysimeter
*
1.6
4.9
6.4
8.2
9.2
8.8
7.4
7.8
4.6
3.0
total,
i nnf\
2.1
2.4
4.5
5.5
8.2
9.8
9.3
8.1
6.9
5.3
3.6
1.8
on£n
Modified Penman §
1975 1976 Avg.
2.8
3.6
4.5
5.8
7.4
9.6
10.8
10.2
8.2
5.4
3.3
2.0
99An
mm/ day
2.4
3.4
4.8
5.9
7.4
9.3
8.8
2.6
3.5
4.6
5.8
7.4
9.4
9.8
10.2
8.2
5.4
3.3
2.0
2201
Calculated from experiment
1975 1976 Avg.
1.8
1.2
4.5
6.7
6.7
6.7
8.5
5.9
4.8
6.2
4.6
1.0
1790
1.7
2.9
3.1
3.9
6.2
8.3
11.9
"
__
1.7
2.0
3.8
5.3
6.4
7.5
10.2
5.9
4.8
6.2
4.6
1.0
1813
mm IBVU zuou ^TU __
*Data taken from Erie et al. (1965). No data given for Ja,,ary or December.
Vta fro, 1975 Annual Research Report
Research Center, ARS, Brawley, Cain.
Calculated by modified
. Wanton. Arizona.
-63-
-------
12
I I
10
a
T3
8
E
E 7
a.
a>
Q
a>
6-
5-
3
2
if Evapo transpiration
J F M A M J JASO
Month of the Year
Figure 22. Comparison of the daily rates of pan evaporation and evapotran-
spiration estimated from the amount of water applied to alfalfa in south-
western Arizona. Average of data from January 1975 to July 1976.
-64-
-------
the ET computed from the amount of water applied. Assuming that the actual
ET for alfalfa is the average of all four measures, the resultant annual ET
is 1991 mm. Corresponding leaching percentages for the 5, 10, and 20% leach-
ing treatments are -3,2, and 10; the flood check had 10% leaching. Data in
later sections tend to confirm that the alfalfa was underirrigated.
Soil Salinity
Initial salinity sensor measurements taken in November 1974 averaged
0.79, 0.62, 0.68, and 0.73 S/m at the 0.3-, 0.6-, 0.9-, and 1.2-m depths,
respectively. Due to operational difficulties with the irrigation system
early in 1975, insufficient water was applied and salinity at the 0.3-m
depth increased. By March, salinity at this depth started to decrease.
This decrease has continued, although short-term cycles have occurred. The
lowest level of salinity (about 0.4 S/m) was reached in January 1976. (See
Fig. 23, and Figs. B-l and B-2 of the appendix.) The salinity below 0.3 m
was expected to increase initially, reaching stable levels after time. Such
increases did occur,, with stable but high levels reached by January 1976.
As shown in Fig. 24, the salinity profiles have changed from ones decreasing
with depth during the first half of 1975 to ones increasing with depth dur-
ing the second half of 1975. They changed little through the first half of
1976, except for a slight increase at 0.9 and 1.2 m. These increases at
greater depths show the beginning of the profile shapes expected to develop,
but no significant salinity difference among treatments has yet been ob-
served. For example, in May 1976, soil salinity at the 1.2-m depth was
about 1.1 S/m for all leaching treatments; for irrigation water having a
salinity of 0.13 S/m, the resultant leaching percentage is 12.
For comparison, the salinity profile for the flood check is shown in
Fig. 24, along with that for the 20% leaching treatment. The salinity in-
creased uniformly with depth, was quite stable, and was 0.2 to 0.3 S/m
lower than in any of the experimental treatments. Based on EC = 0.77 S/m
at the 1.2-m depth, the apparent leaching percentage obtained on the check
plot was 17%, indicating that 5% more leaching had occurred on the flood-
irrigated check than on the sprinkled experimental plots.
Hydraulic Potential
Satisfactory tensiometric data were obtained automatically during a few
periods in the spring and summer of 1975. For instance, hydraulic potential^
profiles for April 1, June 5, and September 15, 1975 are plotted in Fig. 25.
The data for the 10 and 20% leaching treatments were consistent, so the
averages are plotted. The two 5% leaching plots differed widely, however,
so separate plots are given. The reference for the hydraulic potential
values is the soil surface. To obtain matric potentials, the depth must be
added algebraically to the hydraulic potential.
"Hydraulic head is defined as the sum of matric and gravitational
potentials. If both are expressed as equivalent to the height of a column
of water, hydraulic head is obtained by subtracting the depth from the
matric potential read with a tensiometer.
-65-
-------
1.2
c
"o
o
en
i.o H
0.8-
0.6 -
0.4-
0.2-
j L
I i \ i L
10 % Leaching 9.9
X N.
-_;_ ...~.. ....^ ^ .-..
- '6-
Harvest Dates
A A A A A A
A A A A A
I
I
80
160
1975
240
I ]
320
f
40
120
200
1976
] I
280 360
Julian Date
Figure 23. Salinity trends with time for the 10% leaching treatment of the alfalfa experiment
at various soil depths.
-------
0
0.3
0.6
0.9
1.2
1.5
I
0
E
0.3
f 0.6
o> 0.9
O
1.5
Soil Water Salinity , s/m
2 .4 .6 .8 i.o 1.2
T 1 -I
5% Leaching
.2
r~
.4
1.0
1.2
Jan - Jun '75
Jul - Dec '75
Jan - May '76
1.4
.2
A
.6
.8
1.0
1.2
1.4
0.3
0.6
0.9
1.2
I
1.5
flood check
20% Leaching
Figure 24. Time-averaged salinity distributions with depth for three leach-
ing treatments and the flood-irrigation check of the alfalfa experiment.
-67-
-------
0.3
0.6
0.9
1.2
1.5
E 0.3
- 0.6
E 0.9
1.2
1.5
0.3
0,6
0.9
1.2
1.5
a.
UJ
Q
4-1-75
1-05
n-os
i,n -10
n,ir-20
- 6-5-75
9-15-75
-10 -8 -6 -4 -2 0
TOTAL HEAD , m of H20
Figure 25. Hydraulic potential distribution for the three leaching treat-
ments in the alfalfa at selected times during 1975.
-68-
-------
On April 1, the three treatments had similar potential distributions,
with relative values as expected from leaching differences. For all three
treatments, upward gradients occurred in the bottom half of the soil profile
Data for the first replication of the 5% leaching treatment (1-05) are not
available for this date because of equipment problems. By June 5, the pro-
file for 20% leaching had become wetter down to the 1.2-m depth, resulting
in downward water movement to a depth of at least 1.5 m. For 10% leaching,
the wetting was not quite as deep, leaving an upward gradient at the bottom.
In the 11-05 replicate, the top half of the profile was unchanged while the
bottom half was drier. 1-05 was appreciably drier below 1 m than 11-05. By
September 15, the entire profile for 20% leaching had become still wetter;
the profile for 10% leaching continued to wet down to 1.2 m, while drying at
1.5 m erased the upward gradient. The 1-05 replicate was wetter at the
0.6- and 0.9-m levels but had become very dry below that depth, consistent
with the leaching treatments. The 11-05 replicate, however, had become
wetter except for the 0.9-m depth, creating a large upward gradient in the
bottom half of the profile. These data indicate that the soil profile of
11-05 is supplied by water from the groundwater table, which prevents it
from becoming as dry as 1-05. On the other hand, the development of the hy-
draulic profiles in the 10 and 20% leaching treatments suggests that the
upward gradients on April 1 are the result of drying from the top down by
root uptake, rather than upward movement of water from the groundwater
table. The relative positions of the potentials on September 15 are en-
tirely consistent with the presence or absence of drainage from the extrac-
tors and lysimeters at that time, as discussed elsewhere in the report.
The data obtained manually since March 1976 are more reliable. There
are distinct differences within treatments, especially in 1-05. Figure 26
gives hydraulic potential distributions for March 29, May 17, and July 13,
1976. On March 29, the north and south groups of tensiometers in 1-05 gave
similar readings and the average is plotted. As in 1975, 11-05 exhibited
an upward hydraulic potential gradient, which became more pronounced on
May 17. On May 17, the hydraulic potentials in 1-05 indicate that both
sites were being influenced by the water table. Pronounced upward gradients
also existed in 1-05 on July 13. The potential distributions on July 13
represented a period of overirrigation that continued well beyond that date.
All profiles have become much wetter from the top, erasing the upward gradi-
ent in 11-05. The generally heavier irrigations in 1976 kept the profiles
wetter than in 1975 and prevented, for instance, development of a dry layer
at the 0.9-m depth in 11-05 as in 1975 (Fig. 26).
Apparently, the 10 and 20% leaching treatments were generally wet
enough to prevent upward water movement, at least above a depth of 1.5 m.
The f% leaching treatment, however, was sufficiently dry that groundwater
moved upward. The magnitude of any upward fluxes cannot be derived from
tensiometer data, but it seems to vary depending on local subsoil condi-
tions and fluctuations in depth to groundwater.
-69-
-------
0.3
0.6
0.9
1.2
1.5
g 0.3
.0.6
^ 0.9
Q- 1.2
UJ
Q 1.5
0.3
0.6
0.9
1.2
1.5
-10
' 3-29-76 .,
'
*
/ i
i
i
i
/
i i i
i i i
- 5-17-76
* * " * **
fi
V
<- A
x "--.. .'1
x a
- 7-13-76 ' ' ,
- UO- N
- . I-05-S .-' ^
1-05 ^'"
1,11-10 ;*?
H,M- 20,^*'^ .
c ... /
\ -4
i i i
-8 -6 -4
i
// ;
1
I
1
1
1
1
1
y /
/? f
f ' -
]
1
1
.1
/ /^
'^/f ) ~
/ / i
/ 1 /
'//
{ '
[/
\
-2 C
TOTAf. HEAD , m of H£0
Figure 26. Hydraulic potential distribution for the three leaching treat-
ments in the alfalfa at selected times during 1976.
-70-
-------
Vacuum Extractors and Suction Lysimeters
Volume of Leachates
The extractors appear to be working better in the alfalfa than in the
citrus. Pairs of extractors of one replicate generally show good agree-
ment. The same is true for replicates of the same leaching treatment.
While only a few extractors produced some drainage during the summer of
1975, all except extractor I-05-N yielded leachate in the fall. Generally,
the starting dates advanced and the amounts increased with increasing target
leaching, as expected. Interestingly, outflow from all suction lysimeters
started much later than from the corresponding extractors.
TABLE 23 shows the dates of first drainage after January 1, 1975, for
each of the extractors and lysimeters, as well as the total amount of drain-
age until July 26, 1976. The only exceptions are Ext-III-20-N and Ext-III-
20-S, which became defective after July 2, 1976, probably because of a
pressure regulator failure, which allowed the pillow pressure to rise to
over 300 kPa. The amounts of drainage for these two extractors, therefore,
should not be compared directly with those of others. Instead, the leach-
ing percentages should be compared, and they show excellent agreement.
Interpretation of the drainage from the extractors and lysimeters is
complicated by seasonal fluctuations and interruptions of irrigation due to
harvesting. With continuous leaching, higher ET in the summer dictates
wetter soil and more drainage in the summer than in the winter. Nearly all
extractors and lysimeters were working from December 26, 1975 till July 26,
1976. Since this covers approximately half a year from minimum ET to maxi-
mum ET, the comparison shown in the last two columns of TABLE 23 between
replicates over this period appears valid. The measured leaching percent-
ages during this period are about half the target values, except the extrac-
tors in 11-05 which are slightly above target. This appears to be caused by
upward flow from groundwater, as suggested by tensiometric data and other
considerations discussed elsewhere. Generally, the agreement between a
lysimeter and at least one of the corresponding extractors is good. So is
the agreement between replicates of the same leaching treatment. The lower
amount of drainage from Ext-III-20-N occurred primarily before December 26,
1975. After that, the extractor behaved similarly to the other three. The
only extractor that does not appear to work well is II-10-S. The very low
matric potentials in 1-05 suggest that the small leaching volumes obtained
for I-05-N are correct.
Figure 27 shows cumulative leaching percentage for selected extractors
and lysimeters for all three leaching treatments. From TABLE 23, it can
be seen that these are the higher yielding extractors. In November 1975
lysimeter 11-20 started to yield appreciably more than extractors II-20-N
and III-20-S, but since March 1976 the extractors have been draining at a
higher rate. Only the cumulative leaching percentage of lysxmeter 11-20
from late January till the middle of April was above the leaching target.
Similar data for the 10% leaching treatment show that lysimeter 1-10
started yielding drainage about 2 1/2 months later than extractor I-10-S,
-71-
-------
the summer. The leaching percentages are about the same at the end of
July 1976, because they are based on different amounts of irrigation.
Again, only the lysimeter exceeded the target leaching from November till
May.
The extractors in 11-05 behaved nearly identically (Fig. 27). They
started to drain regularly in October 1975, and have done so ever since,
mostly at a leaching percentage above the target value. In contrast, the
extractors in 1-05 yielded far less and behaved qualitatively more like
those in the higher leaching treatments. The lysimeter in 11-05 started to
drain late in December 1975 and was above target till early May 1975, but
has not yielded much since. Appendix Figs. B-3, B-4, and B-5 give the
TABLE 23. DRAINAGE FOR EACH VACUUM EXTRACTOR AND SUCTION LYSIMETER IN THE
ALFALFA EXPERIMENT AFTER JANUARY 1, 1975 UNTIL JULY 26, 1976
Repli-
cation
Ext-I-05-N
Ext-I-05-S
Lys-I-05
Ext-II-05-N
Ext-II-05-S
Lys-II-05
Ext-I-10-N
Ext-I-10-S
Lys-I-10
Ext-II-10-N
Ext-II-10-S
Lys-II-10
Ext-II-20-N
Ext-II-20-S
Lys-II-20
Ext-III-20-N
Ext-III-20-S
Lys-III-20
Date Total drainage
first to
drainage 07-26-76
03-25-76
09-03-75
01-06-76
09-18-76
09-03-75
12-26-75
09-03-75
06-02-75
12-05-75
06-02-75
10-06-75
12-15-75
09-03-75
06-02-75
09-18-75
06-02-75
06-02-75
09-03-75
mm
14.5
28.3
17.2
82.4
85.1
26.5
72.4
97.2
53.6
63.7
24.0
61.6
213.6
178.0
220.7
124.0*
167.5*
206.5
Total drainage
between 12-26-75
and 07-26-76
mm
10.2
24.0
17.2
62.7
68.8
23.6
43.6
61.1
40.0
48.1
20.7
52.4
140.0
110.1
127.6
87.9*
95.8*
105.6
% leaching be-
tween 12-26-75
and 07-26-76
0.9
2.1
1.5
5.5
6.1
2.1
3.6
5.0
3.3
4.0
1.7
4.3
10.5
8.2
9.6
8.9*
9.7*
7.9
Until July 2, 1976.
-72-
-------
SEP NOV JAN MAR MAY JUL
OCT DEC FEB APR JUN
975
1976
Figure 27. Accumulated leaching percentages based on leachate from vacuum
extractors and suction lysimeters for the three
alfalfa leaching treatments.
-73-
-------
cumulative irrigation and drainage, as well as the cumulative drainage per-
centages of the various lysimeters and extractors, allowing a more detailed
analysis.
With the exception of 11-05, drainage per unit area from the lysimeters
tended to be about the same as from the extractors, but it was faster over
shorter time intervals. This suggests the suction in the ceramic tubes of
the lysimeters was maintained too high. This could cause them to yield too
much drainage compared with outside the lysimeters if the soil were dried
out so much that it reduced the ET compared to the ET outside. There are
no tensiometers in the lysimeters to check this. If wet, the lysimeters
work under a negative feedback principle; if the ceramic tubes cannot handle
the drainage flow passing by, the water will back up against the bottom of
the lysimeters, which will wet up the soil around the tubes, increase the
hydraulic conductivity, and increase the drainage rate. The situation with
the extractors is different. If they cannot handle the drainage flow, the
water will pass by without a chance to make up for it. If the suction in
the ceramic tubes of the extractors -is maintained too high, they could
pull in too much flow. Actually, however, the suctions in the ceramic tubes
of the extractors were regulated such that the tensiometers immediately
above and adjacent to the extractors read about the same.
In view of the above consideration, it is likely that, if there is a
discrepancy between the extractors and lysimeters, the latter will be high.
There is one notable exception to this prediction: both extractors in 11-05
yielded almost three times as much drainage per unit area as the lysimeter.
Tensiometric data indicate this was the result of upward flow from the water
table. For instance, Figs. 25 and 26 show an upward hydraulic potential
gradient fci. the 11-05 replicate. Of all replicates, only 11-05 showed con-
sistently upward gradients. Also, the total hydraulic potentials in 11-05
were much higher than in 1-05, about -1.5 m versus -5.0 m, and were even
higher than in the 10% leaching treatments. Only during July 1976, when
the field was overirrigated, was the hydraulic gradient reversed by wetting
from above. The upward flow could not enter the lysimeter in 11-05. Thus,
the contrast between the sustained drainage of the extractors and the
behavior of the lysimeter, which was more in line with that of the others,
is entirely compatible with upward flow.
In summary, the extractors and lysimeters indicate that the alfalfa has
been underirrigated at about half the values of each leaching target. The
extractors in 1-05 are affected by upward flow, and the suctions in the
lysimeters should probably be lowered.
Composition of Leachates
The waters collected by the extractors are analyzed to ascertain the
amounts and compositions of the salt loss by deep percolation. To date, 137
samples have been analyzed. Compositions of drainage waters samples col-
lected in April 1976 are given in TABLE 24. Comparisons of compositions
with time are given in TABLE 25.
The salinity of the drainage water has increased with time. The earli-
est samples reflect previous management; the later samples show higher
-74-
-------
TABLE 24. COMPOSITIONS OF DRAINAGE WATERS FOR ALFALFA FIELD, APRIL 1976
Extractor
No.
pH
Ca
Mg
Na K Sum of HCO
cations
»eq/A
so4
Cl
N03
EC*
S/m
SARt
Leaching
fraction
Cl. /Cl,
iw dw
5% Leaching
1-5 -NJ
I-5-S
II-5-N
II-5-S
7.0
6.9
6.7
7.0
46.6
35.7
32.2
37.7
32.0
22.6
44.0
37.3
37.6 1.3
34.8 1.1
140.0 0.9
91.4 0.8
118
94
217
167
16.4
13.8
25.0
25.3
48.1
38.6
137.0
67.5
52.6
43.4
56.6
73.8
0.2
0,2
0.2
0.2
0.87
0.75
1.46
1.23
6.0
6.4
23.0
15.0
Or\f\
. uo
10% Leaching
I-10-N
I-10-S
H-10-N
II-10-S
6.9
7.4
6.6
6.8
54.0
47.1
33.0
43.4
41.8
40.8
36.4
34.2
81.0 0.9
121.0 1.0
21.8 0.4
32.7 0.6
178
210
92
111
19.5
23.7
20.5
22.1
81.7
113.0
44.9
52.4
75.8
77.8
32.3
40.0
0.4
0.3
0.4
0.2
1.24
1.43
0.68
0.79
12.0
18.0
3.7
5.2
0.06
20% Leaching
II-20-N
II-20-S
III-20-N
III-20-S
6.8
6.8
6.4
6.3
42.0
50.2
34.4
52.7
34.5
43.2
20.7
35.2
66.8 0.4
39.4 0.5
26.4 0.5
32.0 0.3
144
133
82
120
25.4
20.1
21.7
19.0
77.9
74.6
35.9
71.3
40.3
40.3
23.5
37.8
0.1
0.1
0.1
0.1
0.98
0.88
0.60
0.83
11.0
5.8
5.0
4.8
0.09
Electrical conductivity at 25°C.
SAR = Na//t(Ca + Mg)/2], where solutes are in meq/liter.
A
I-5-N denotes the extractor in the north position of the first replication in the 5% leaching
treatment.
-------
TABLE 25,
Extractor
No.
%
I-5-S§
II-5-N
I-10-S
II-10-N
II-20-S
III-20-S
Date
12-74
04-76
12-74
04-76
12-74
06-75
04-76
12-74
06-75
04-76
12-74
06-75
04-76
12-74
06-75
04-76
. CHANGES IN
Ca
33.5
35.7
29.1
32.2
32.5
61.4
47.1
14.4
11.4
33.0
20.8
26.8
50.2
20.8
25.4
53.8
Mg
17.3
22.6
23.0
44.0
21.0
46.4
40.8
14.9
17.7
36.7
17.0
22.7
43.2
11.1
14.7
35.8
COMPOSITIONS
Na
27.5
34.8
62.7
140.0
43.1
81.0
121.0
18.2
18.5
23.0
22.3
24.9
39.4
22.8
23.4
33.2
K
0.9
1.1
0.6
0.9
0.6
1.0
1.0
0.3
0.3
0.4
0.3
0.4
0.5
0.4
0.4
0.5
OF DRAINAGE WATERS OF
Sum of
cations
/"*>*
79.0
94.3
115.0
217.0
97.2
190.0
210.0
47.8
47.9
93.1
60.3
74.7
133.0
55.1
64.0
123.0
HC03
10.6
13.8
12.6
25.0
12.2
20.2
23.7
15.0
14.2
20.5
11.5
18.0
20.1
11.0
18.0
18.9
so4
38.7
38.6
75.4
137.0
54.1
97.8
113.0
19.6
23^3
44.9
35.0
43.7
74.6
28.7
30.9
71.3
ALFALFA FIELD WITH TIME
Cl
25.1
43.4
28.1
56.6
32.2
69.6
77.8
11.1
11.4
32.3
14.1
15.3
40.3
14.6
15.9
37.8
N03
2.0
0.2
0.4
0.2
2.3
0.7
0.3
0.4
0.2
0.1
0.3
0.3
0.1
0.8
0.3
0.1
sio2
mg/A
56
48
67
38
55
57
37
58
57
37
61
66
43
23
48
43
EC*
S/m
0.59
0.75
0.83
1.46
0.74
1.23
1.43
0.38
0.37
0.68
0.46
0.54
0.88
0.42
0.48
0.83
SARt
5.5
6.4
12.0
23.0
8.3
11.0
18.0
4.8
4.8
3.9
5.1
5.0
5.8
5.7
5.2
5.0
Electrical conductivity at 25°C.
SAR = Na/v/KCa + Mg)/2], where solutes are in meq/liter.
»
I-5-S denotes the extractor in the south position of the first replication in the 5% leaching
treatment.
-------
values Because the imposed treatments result in lower leaching fractions.
While in all cases the concentrations have increased, the present composi-
tes are not well related to treatments, nor are replicates very similar.
Drainage waters collected from extractors 1-10 are higher in all major salt
constituents than waters collected from extractors 1-05. Waters collected
from extractors 11-10 are much lower in salts than their replicates (1-10)
or the waters collected from extractors 11-20. Nevertheless, the average
leaching percentages for the three treatments given in TABLE 24 tend to
support the leachate volume data shown in Fig. 27. By April 1976, the
accumulated leaching percentages on a drainage volume basis were 7, 10
and 16 for the 5, 10, and 20% leaching treatments.
Predicted drainage water compositions as a function of leaching are
given in TABLE 26. The present compositions do not yet correspond well with
the predicted values. For a given chloride concentration, the present con-
centrations of magnesium, sodium, and sulfate are lower than expected, and
calcium is higher. Insufficient time has elapsed since the experiment was
initiated to attain steady state; continuing exchange reactions account for
the departure in composition from predicted levels, including EC.
Yield
The yields for the three leaching treatments and the flood check are
given for each replication in TABLES 27 and 28 for 1975 and 1976, respec-
tively. Total yields for 1975 were 17.9, 17.6, and 17.6 Mg/ha for the 5,
10, and 20% leaching treatments, respectively. Likewise, yields for 1976
were 16.4, 15.8, and 15.9 Mg/ha for 5, 10, and 20%. The yield data indicate
no significant differences among the leaching treatments. Total yields for
1976 were lower than for 1975 because fewer cuttings were taken, although
yields by cutting were consistently higher in 1976. As expected, the yields
of all treatments gradually increased with each cutting until early summer
as the density of the alfalfa stand increased during the spring each year
and then decreased during late summer.
Encroachment of grass and other weeds has plagued the experiment, and
the flood check has had consistently higher yields except in midsummer when
the flood check was probably underirrigated. Annual flood-check yields were
21.3 and 18.6 Mg/ha for 1975 and 1976, respectively -- 20 and 16% larger
than the average experimental yields. After the sixth cutting in 1975, parts
of the experimental area were cultivated lightly and reseeded in an attempt
to eliminate weeds. This accounts for the discrepancy in cutting dates and
resultant yields for the seventh cutting in 1975 (see TABLE 27). After the
fifth cutting in 1976, the alfalfa was completely removed from the field to
eliminate weeds. The field was allowed to dry throughout August and Septem-
ber to kill the grass and then was cultivated 0.2 m deep and leveled with a
laser plane. The field was replanted the first of October.
Leaf Analysis
Alfalfa shoots were sampled in September 1975 to evaluate plant nutri-
ent status. Samples were taken before the sixth harvest from five locations
randomlj selected along the length of each experimental plot and along three
-77-
-------
TABLE 26. PREDICTED DRAINAGE WATER COMPOSITIONS FOR THE ALFALFA TREATMENTS
oo
I
Leaching
fraction
.03
.05
.10
.15
.20
.30
.40
Ca
24.8
26.2
28.4
22.9
19.9
16.8
15.2
Mg
90.0
54.0
27.0
18.0
13.5
9.0
6.7
Na
227.0
136.0
68.0
45.3
34.0
22.7
17.0
Sum of
cations
/ o
Tneq/JG
341.0
216.0
123.0
86.2
67.4
48.5
39.0
HC03
18.6
15.0
11.8
11.6
11.4
11.1
11.0
S°4
206.0
131.0
76.6
51.3
38.5
25.7
19.2
Cl
117.0
70.0
35.0
23.3
17.5
11.7
8.7
ECt
S/m
2.51
1.68
0.98
0.72
0.56
0.39
0.32
SAR^
29.9
21.5
12.9
10.0
8.3
6.3
5.1
Assuming the partial pressure of C02 is 0.080 and the irrigation water composition is, in meq/£:
Ca 4.5, Mg 2.7, Na 6.8, Cl 3.5, HCO, 2.8, and SO 7.7.
t
Electrical conductivity at 25°C.
SAR = Na//[(Ca + Mg)/2], where solutes are in meq/liter.
-------
TABLE 27. ALFALFA YIELD (Mg/ha)* FOR EACH REPLICATION OF THE
LEACHING TREATMENTS AND FOR THREE LOCATIONS WITHIN
THE FLOOD CHECK FOR THE 1975 SEASON
Leaching
treatment
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
7th
5%
10%
20%
Flood
Replication
1
1.0
1.1
0.9
2.2
2.6
2.3
2.9
2.7
3.0
3.1
3.1
3.5
3.3
3.6
3.4
3.6
3.6
3.3
3.9
3.5
3.8
3.2
2.9
3.3
Cutting
1.7
2.0
1.8
2.7
2 3
1st Cutting
1.1 1.0
0.9 1.2
1.0 1.2
2.1 2.4
2nd Cutting
2.2 2.3
2.3 2.2
2.2 1.9
2.5 2.3
3rd Cutting
2.8 2.9
2.7 2.8
3.0 2.6
4.0 3.8
4th Cutting
3.3 3.0
3.4 3.3
3.5 3.2
3.2 3.8
5th Cutting
3.4 3.2
3.5 3.6
3.4 3.3
3.5 2.8
6th Cutting
3.1 3.0
3.2 3.3
3 0
J \J
3.3 3.3
flood cut 11-24-75
1.6 1-5
1.5 1-7
1-7 1 R
1.7 -L'8
0 c ? fi
2 . -> *"°
4
02-24-75
1.5
1.4
1.3
04-30-75
2.0
1.9
2.2
06-04-75
2.8
2.7
2.8
07-04-75
3.2
3.1
3.0
08-11-75
3.9
3.1
3.8
09-17-75
3.0
3.3
3.4
experiment
1.8
1.8
1.8
5
1.2
1.1
0.9
2.1
1.7
2.0
3.3
2.6
2.6
3.2
3.4
3.1
4.0
3.8
3.7
3.0
2.9
3.2
Average
yield
1.2
1.2
1.0
2.3
2.2
2.1
2.2
2.5
2.9
2.8
2.8
3.8
3.2
3.3
3.2
3.5
3.6
3.4
3.6
0 0
J . J
3.2
3.2
3.1
3.3
cut 01-27-76
1.3 1.6
0.7 1.6
1.6 1.7
2.6
*Note: 1 Mg/ha = 0.466 ton/acre.
-79-
-------
TABLE 28. ALFALFA YIELD (Mg/ha)* FOR EACH REPLICATION OF THE
LEACHING TREATMENTS AND FOR THREE LOCATIONS WITHIN THE
FLOOD CHECK FOR THE FIRST FIVE CUTTINGS OF 1976
Leaching
treatment
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
5%
10%
20%
Flood
Replication
1
1.5
1.5
1.5
2.6
3.7
3.3
3.5
4.2
4.1
3.6
4.2
4.3
4.0
3.9
3.5
4.4
4.3
3.9
4.3
3.8
2 3
1st Cutting
1.5 1.5
1.8 1.3
1.7 1.3
2.4 2.3
2nd Cutting
3.3 3.7
3.6 3.3
3.5 3.2
4.0 4.1
3rd Cutting
3.9 4.0
3.8 4.1
3.4 3.8
3.8 4.8
4th Cutting
3.4 3.8
3.8 3.9
3.7 3.4
3.9 3.7
5th Cutting
3.8 3.7
4.4 3.9
4.4 3.7
3.6 3.8
4
03-09-76
1.3
1.2
1.2
04-23-76
3.2
3.3
3.4
05-26-76
4.0
3.4
3.6
06-23-76
3.5
3.6
3.6
07-30-76
3.8
4.3
3.9
5
1.1
1.0
1.1
3.7
3.2
3.3
3.8
3.6
3.6
3.6
3.2
3.5
3.7
2.7
3.1
Average
yield
1.4
1.4
1.4
2.4
3.5
3.3
3.4
4.1
3.9
3.7
3.7
4.3
3.7
3.6
3.5
4.0
3.9
3.8
3.9
3.8
Note: 1 Mg/ha = 0.446 ton/acre.
-80-
-------
iengttu? wirliin the flaofi check. Each sample consisted of approximately 200 g
fresh weight of plants, cut about 0.05 m above the soil. The samples were
washed, oven-dried, and analyzed for 13 mineral elements (TABLE 29). Com-
pared with published tissue analyses used to evaluate nutrient status (Soil
Testing anc Plant Analysis, Part H, p. 79, Soil Sci. Soc. Amer., 1967), Ca
levels appear low; Ha and S are somewhat high. All other elements are within
normal concentration ranges. Hie Ma analyses indicate that foliar absorption
may be responsible for the high Na levels in the experimental treatments,
but, surprisingly, Replicate 5 in each leaching treatment had significantly
lower Ha levels than the other replications. Besides Na, significant, but
small, differences in mineral contents between the flood check and experi-
mental plants were present for K, Ca, and Cl.
TABLE 29. MIHERAL COMPOSITION OF ALFALFA
Leaching Treatment
Element. {unit}
* ff>
P mmoIes/lOOg
F. meq/lOOg
f- ' '*
Kg
Us
C-L. "
£ "
B u-g/g
Je
Mn "
Zn
'*"*-( ^ "
5%
3.2
7.
66
67
25
25*
35
30
66
90
25
26
13
WZ
3.2
7.6
69
66
24
24*
34
29
62
90
23
23
13
20%
3.2
7.9
68
65
25
27*
33
29
64
87
25
22
13
Flood
check
3.2
8.0
82
73
23
6
25
27
65
85
27
26
13
*Average of replicates 1-4; Eeplicate 5 in all treatments was signifi-
cantly lower {average 10 meq/100 g).
Water Table
During the installation of the manholes in September 1974, the water
n 2 m of the soil surface. This was caused in part, by the
:f the experiment.
One well
four comers
-81-
-------
of the leaching experiment field, and one at about the midpoint of each side
of the field. Depth to water table below ground elevation is given in TABLE
30. To date, the depths to the water table have ranged from 2.0 to 3.5 m.
From observations during drilling, a sand stratum seems to be present at the
depths where water is found.
TABLE 30. DEPTH TO WATER TABLE IN ALFALFA EXPERIMENT (m)
Date
05-21-75
08-08-75
09-18-75
10-31-75
12-05-75
01-06-76
02-05-76
03-05-76
04-05-76
05-05-76
06-08-76
07-16-76
08-20-76
NW
2.6
3.3
3.1
2.9
2.7
>5.0
2.5
2.3
2.0
2.3
2.0
2.4
NE
2.5
>5.0
>5.0
2.5
2.5
2.7
>5.0
>5.0
>5.0
>5.0
>5.0
>5.0
>5.0
Position*
CW
>5
3.2
3.1
3.0
2.9
3.5
3.5
3.1
3.1
2.8
2.8
>5.0
3.0
CE
2.9
2.9
2.9
2.9
2.6
2.8
2.8
2.9
2.9
2.6
2.9
2.8
2.7
SW
>5.0
>5.0
2.6
2.8
2.9
2.9
2.7
3.1
3.0
2.7
2.8
2.8
>5.0
SE
2.9
>5.0
3.1
2.7
2.9
2.5
2.7
2.5
2.5
2.5
2.2
2.2
2.2
*
CW refers to center of field on the west, CE to center of field on the
east; the remaining locations correspond roughly to field corners, i.e.,
(NW) northwest, (NE) northeast; (SW) southwest, and (SE) southeast.
Soil Air
Soil Oxygen
In conjunction with the soil oxygen measurements in the citrus experi-
ment, Dr. Burl Meek (ARS, Brawley, Calif.) made similar measurements in the
alfalfa at a soil depth of 0.45 m in two of the five replications in each of
the three leaching treatments and at several locations in the flood check.
Measurements were made for several days in succession three different times
the first half of 1975. TABLE 31 shows the average soil oxygen content for
each treatment.
In March, when the average irrigation rate was about 5 mm/day in the
experimental plots, the soil oxygen content was well above 10%. During
April 1975, about 10 mm/day of irrigation water was being applied to decrease
the soil salinity to the desired level after some operational problems with
the irrigation system. Soil oxygen levels during this period were well be-
low 10% for the 20% leaching treatment and near 10% for 10% leaching. Visual
-82-
-------
TABLE 31. PERCENT SOIL OXYGEN AT THE 0.45-m SOIL DEPTH
IN THE ALFALFA EXPERIMENT
Date
March 1975
April 1975
June 1975
Leaching treatment-
5%
17.5
16.0
17.1
10%
18.6
10.5
16.3
20%
18.2
5.8
14.0
Flood
14.1
11.1
effects of low soil oxygen were apparent in some plots. The alfalfa in rep-
lication 3 of the 20% leaching treatment was yellowish green, whereas the
plants in replication 2 appeared normal. Oxygen for replication 3 averaged
2.4%; the average for replication 2 was 12.0%. Similarly, the 10% leaching
plot of replication 2 had a soil oxygen content of 5.9% and the alfalfa had
a light-yellow tint, while replication 1 appeared normal with an oxygen con-
tent of 13.9%. Because of the drastic decrease in soil oxygen, the irriga-
tion rate was reduced, thus extending the period needed to decrease the soil
salinity. In June, the average irrigation rate was about 9 mm/day, but since
ET was also high, soil oxygen was well above 10%.
Soil Carbon Dioxide
Carbon dioxide concentration in soil air varies with time of year and
soil depth as shown in TABLE 32. The values for alfalfa are about two to
four times greater than the concentrations obtained in the citrus (see TABLE
19) . The higher values for alfalfa are reasonable because the entire soil
surface is irrigated, whereas only about one-half the area is irrigated in
the orange grove. The soil in the alfalfa field is also finer textured.
Both factors tend to reduce gaseous exchange between the soil and atmosphere
in the alfalfa. The carbon dioxide concentrations in the flood check are
similar to those in the leaching plots. Note that the carbon dioxide concen-
trations are highest immediately after a flood irrigation, which corresponds
to the time the soil is wettest.
The amount of salt in the drainage water depends on leaching fraction
and the carbon dioxide in the soil air. As leaching decreases, however,
the influence of carbon dioxide on the salt load decreases for Colorado
River water (Oster and Rhoades, 1975). For leaching fractions of 0.2 and
0.1 and with 10% carbon dioxide in the soil air, as under alfalfa, the res-
pective predicted salt loads are 4.6 and 8.1 kg of salt per m of drainage
water. Comparable numbers for 3% carbon dioxide, as in the citrus, are 4.2
and 7.8 kg/m3. The decrease is a direct result of the decreased solu-
bility of lime with decreased carbon dixoide level. The relative reduction
is 9 and 4% for leaching fractions of 0.2 and 0.1. At leaching fractions
<0.1, the salt that can be dissolved in the drainage water resulting from
irrigation with Colorado River water also depends on the solubility of
gypsum. Decreases in the solubility of lime result in increased solubility
of gypsum and vice versa.
-83-
-------
TABLE 32. AVERAGE PERCENT CARBON DIOXIDE IN SOIL AIR
UNDER ALFALFA AS A FUNCTION OF SOIL DEPTH
Soil
depth
Oct 75
Leaching Experiment
Jan 76
Apr 76
Jul 76
meter
0.3
0.6
0.9
1.2
2.3
4.7
7.7
8.5
4.0
5.3
6.5
7.2
5.5
6.1
7.8
8.2
4.2
9.3
9.7
10.1
Soil
depth
Sep 75
Flood Check
Jan 76
Apr 76
August 1976
18*
16*
21*
4*
7*
meter
0.3
0.6
0.9
1.2
1.5
2.2
5.4
7.1
7.9
8.9
1.7
4.2
5.0
5.7
6.8
4.1
6.9
9.7
9.3
9.6
6.8
9.4
11.6
12.1
12.6
3.7
7.3
9.7
10.6
12.1
Days after irrigation.
DISCUSSION
The results presented for alfalfa are highly tentative and several addi-
tional years of experimentation are required for substantiation. It appears
that a high irrigation efficiency can be attained with the level-basin,
flood-irrigation technique. It does not appear that irrigation efficiency
can be improved significantly for alfalfa by a change in irrigation method
because of its deep rooting nature. Leaching percentage for the flood check
was about 10%.
Evapotranspiration
Following the procedure used for the citrus, ET can be estimated by
correcting the depth of irrigation and rainfall by the design leaching frac-
tions, by calculations based on meteorological data, and by estimates based
on the volumes or chloride concentrations of leachate from the vacuum ex-
tractors and suction lysimeters. These estimates of ET can also be compared
with published values.
The estimate of annual ET based on water application and the desired
leaching during the past 19 months is 1813 mm. Calculations from the
modified Penman equation result in an annual ET estimate of 2201 mm. Erie
et al. (1965) published a value of 1890 mm for alfalfa near Phoenix,
Arizona, and lysimeter measurements at Brawley, California, indicate a
yearly ET of 2060 mm.
-84-
-------
Leaching percentages based on both volume (Fig. 27) and chloride
tration (TMLE 24) of leachates from the vacuum extractors and s~uSion
S^f ^die^te the sprinkled treatments were under-irrigated. If an annual
ET of about 2000 mm were postulated, based on the estimates noted above, it
would also indicate underirrigation.
Salinity Trends and Leachates
The salinity profiles developed slowly, as expected, and probably still
have not reached equilibrium. The shapes of the profiles by mid-1976 indi-
cated an apparent depth of rooting of 0.90 m for the 5% leaching treatment
and a rooting depth in excess of 1.20 a for the other treatments. The flood
check data indicate a probable rooting depth substantially deeper than
alfalfa on the sprinkled treatments.
Based on the volumes of water extracted, leaching percentages for the
three treatments are estimated to be about 34 5, and 10%; based on composi-
tion of the leaching water, they are 6, 6, and 9%; based on the salt
sensors, the leaching percentages at the 1.20-m depth for all three treat-
ments are about 12%. One may conclude that the leaching fractions obtained
to date were less than those planned; this observation is consistent with
the water application data and the ET estimates.
Crop Yields
Yields from the leaching experiment have been consistently lower than
those from the flood check. The frequent application of small amounts of
water .resulted in a. high soil matric potential at shallow depth, which
apparently stimulated weed growth. The high soil water content during
harvest aggravated soil compaction by heavy equipment. Infiltration rates
"Here low, again causing an unfavorable soil surface condition. Early in
the experimental period, an excess of herbicides was applied accidentally
to the plots. All these factors combined resulted in a rather poor
alfalfa stand, compared to that on the flood check. Whatever the reason,
the time-averaged in situ salinity in the 0- to 1.20-m root zone for the
period July 1975 to~May 1976 was about 0.5 S/m for the flood check com-
pared to 0.8 S/m for the three experimental treatments. The yield depression
expected from this difference (Maas and Hoffman, 1977) is 12%; the actual
reduction observed (33.5 vs. 39.9 Ife for the first 12 cuttings) was about
16%.
Overview
The data suggest, but do not definitely establish, that the experi-
mental plots were underirrigated more than was the flood check, notwith-
standing comparable application rates. Quite possibly, the flood check
used water from the water table, further confounding the water-use picture.
This would explain, in part, the substantial difference in the salt pro-
files (Jig 24) for identical water application rates. It as also possible
that higher evaporation from the exposed soil surface following each
-85-
-------
cutting decreased the water available for leaching in the sprinkled treat-
ments. Plainly, it has been difficult, with the irrigation system used, to
obtain water infiltration rates consistent with projected needs.
The alfalfa was reseeded in the fall of 1976 and the irrigation system
has been modified by installing larger capacity nozzles. It is expected
that somewhat less frequent, but larger, irrigation applications will sim-
plify the management and permit the soil to dry more before each cutting,
thereby overcoming the severe weed and soil compaction problems associated
with an excessively wet soil surface.
-86-
-------
REFERENCES
Austin, R. S., and S. L. Rawlins. 1977. Photo-interrupter liquid-level
detector circuit: Applications for automatic irrigation with tensiom-
eters. Agr. Eng. (submitted)
Bernstein, L., and L. E. Francois. 1973. Leaching requirement studies:
Sensitivity of alfalfa to salinity of irrigation and drainage waters.
Soil Sci. Soc. Amer Proc. 37:931-943.
Bessler, M. B., and J. T. Maletic. 1975. Salinity control and federal
water quality act. J. Hydraulics Div., ASCE 101(HY5): 581-594.
Bingham, F. T., R. J. Mahler, J. Parra, and L. H. Stolzy. 1974. Long-term
effects of irrigation-salinity management on a Valencia orange orchard.
Soil Sci. 117:369-377.
Chapman, H. D. 1967. Plant analysis values suggestive of nutrient status
of selected crops. In; Soil Testing and Plant Analysis. Part II,
Plant analysis. Soil Sci. Soc. Amer. Special Publ. Series No. 2,
p. 79, Madison, Wisconsin.
Doorenbos, J., and W. 0. Pruitt. 1975. Crop water requirements. Irriga-
and Drainage Paper No. 24, Food and Agric. Organization of the U.N.,
Rome, 179 pp.
Duke, H. R., and H. R. Haise. 1973. Vacuum extractors to assess deep
percolation losses and chemical constituents of soil water. Soil Sci.
Soc. Amer. Proc. 37:963-964.
Erie, L. J., 0. F. French, and K. Harris. 1965. Consumptive use of water
by crops in Arizona. Univ. Arizona Agr. Exp. Sta. Tech. Bull.
169, 44 p.
Jensen, M. E. (ed). 1973. Consumptive use of water and irrigation water
requirements. Report of the Tech. Comm. on Irrig. Water Requirements
of the Irrig. and Drainage Div., Amer. Soc. Civil Engr., 215 pp.
Jones, W. W., T. W. Embleton, and C. B. Cree. 1957. Number of replications
and plot sizes required for reliable evaluation of nutritional studies
and yield relationships with citrus and avocado. Amer. Soc. Hort.
Sci. 69:208-216.
LeMert R. D., and M. T. Kaddah. 1977. Lysimeter-determined and estimated
evapotranspiration of alfalfa in the arid subtropical climate of
Imperial Valley, California. Agron. J. (in preparation)
-87-
-------
Maas, E. V., and G. J. Hoffman. 1977. Crop salt tolerance - Current as-
sessment. J. Irrig. and Drainage Div., Proc. Amer. Soc. Civil Eng.
(in press)
Oster, J. D., and J. D. Rhoades. 1975. Calculated drainage water composi-
tions and salt burdens resulting from irrigation with river waters in
the Western United States. J. Environ. Qual. 4:73-79-
Oster, J. D., and L. S. Willardson. 1971. Reliability of salinity sensors
for the management of soil salinity. Agron. J. 63:695-698.
Oster, J. D., L. S. Willardson, J. van Schilfgaarde, and J. 0. Goertzen.
1976. Irrigation control using tensiometers and salinity sensors.
Transactions of the ASAE 19:294-298.
Parker, E. R., and L. D. Batchelder. 1932. Variation in the yield of
fruit trees in relation to the planning of future experiments.
Hilgardia 7(2):81-161.
Raats, P. A. C. 1975. Distribution of salts in the root zone. J. Hydrol.
27:237-248.
Reisenauer, H. M. (ed). 1976. Soil and plant-tissue testing in California.
Div. of Agric. Sci., University of California Bull. 1879, 54 pp.
Replogle, J. A. 1977. Portable, adjustable flow-measuring flume for small
canals. Transactions of the ASAE (in preparation)
Rhoades, J. D., J. D. Oster, R. D. Ingvalson, J. M. Tucker, and M. Clark.
1974. Minimizing the salt burdens of irrigation drainage waters.
J. Environ. Qual. 3:311-316.
Richards, L. A. 1966. A soil salinity sensor of improved design. Soil
Sci. Soc. Amer. Proc. 30:333-337.
United States Environmental Protection Agency. 1971. The mineral quality
problem in the Colorado River Basin. Summary Report, U.S. Environ-
mental Protection Agency Regions 8 and 9, GPO 790485.
Van Schilfgaarde, J., L. Bernstein, J. D. Rhoades, and S. L. Rawlins. 1974.
Irrigation management for salt control. J. Irrig. and Drainage Div.,
ASCE 100(IR3): 321-338.
-88-
-------
TABLE A-l. DESCRIPTION OF SOIL AT THE CITRUS EXPERIMENTAL SITE
PROVIDED BY THE SOIL CONSERVATION SERVICE
Date: September 27, 1973. Area; Yuma County, Arizona; Lower Gila River Area - 649.
Description by: R.L.A.
Classification; Typic Haplargid, coarse-loamy, mixed, hyperthermic.
Location: Between 21st & 22nd tree west of road & between 19th and 20th tree north of road in East
Block No. 220D16, Sec. 20, R. 16 W., T. 8 S.
Vegetation; Valencia oranges. Climate; Arid. Parent Material: Old alluvium. Physiography; Mesa.
Relief; A-nearly level. Elevation: 325 feet. Slope; Less than 1%. Aspect: North. Erosion;
None to slight. Permeability: Moderate. Drainage; Well. Ground Water; Deep. Moisture; Moist
to depth described. Root distribution; Normal.
Additional Notes; Coarse gravel found in a sand matrix from 144 to 206 in. (Dateland fine sandy loam.)
HORIZON DESCRIPTION
Ap 0-6 inches (0-15 cm), brown (10YR 5/3) fine sandy loam, dark brown (10YR 4/3) moist; massive
oo structure; soft, very friable, slightly sticky, nonplastic; many fine and very fine roots, many
f fine and very fine tubular pores; slightly effervescent; moderately alkaline (pH 8.2); abrupt
smooth boundary.
B2t 6-17 inches (15-43 cm), light yellowish brown (10YR 6/4) fine sandy loam, dark yellowish brown
(10YR 4/4) moist; massive structure; soft to slightly hard, very friable, sticky, slightly
plastic; common fine and very fine and very few medium roots, common fine and very fine tubu-
lar pores; few thin clay bridges; strongly effervescent; moderately alkaline (pH 8.2), clear
smooth boundary.
B3t 17-27 inches (43-69 cm), light yellowish brown (10YR 6/4) fine sandy loam, dark yellowish
brown (10YR 4/4) moist; massive structure; soft, very friable, sticky, slightly plastic; very
few medium and common fine and very fine roots, many fine and very fine tubular pores; few
thin clay bridges; strongly effervescent, fine, irregularly shaped, pinkish white soft masses
of lime (2% by volume); moderately alkaline (pH 8.2), clear smooth boundary.
Clca 27-54 inches (69-137 cm), pale brown (10YR 6/3) loam, dark brown (10YR 4/3) moist; massive
structure; soft, very friable, slightly sticky, slightly plastic; very few medium and coarse
and common fine and very fine roots, common fine and very fine tubular pores; violently ef-
-------
HORIZON DESCRIPTION
fervescent, fine and medium, irregularly shaped, pinkish white soft masses of lime (25% by vol-
ume); moderately alkaline (pH 8.2) clear wavy boundary.
C2 54-57 inches (137-145 cm), pale brown (10YR 6/3) fine sandy loam, dark brown (10YR 4/3) moist;
massive structure; soft, very friable, slightly sticky, nonplastic; common very fine and few
fine roots, many fine and very fine tubular pores; strongly effervescent, fine irregularly
shaped concretions of lime (<1% by volume); moderately alkaline (pH 8.2), clear wavy
boundary.
i
o IIC3 57-74 (145-188 cm), pale brown (10YR 6/3) sand, brown (10YR 5/3) moist; massive structure;
loose, very friable, nonsticky, nonplastic; very few very fine roots, many interstitial
pores; noneffervescent, moderately alkaline (pH 8.2); cleary wavy boundary.
IIC4 74-84 inches (188-214 cm) pale brown (10YR 6/3) sand, dark brown (10YR 4/3) moist, massive
structure; loose, very friable, nonsticky, nonplastic; many interstitial pores; 15% gravel (by
volume), noneffervescent, moderately alkaline (pH 8.2), clear wavy boundary.
IIC5 84-94 inches (214-239 cm), pale brown (10YR 6/3) sand, dark grayish brown (10YR 4/2) moist;
massive structure; loose, very friable, nonsticky, nonplastic, many interstitial pores; noneffer-
vescent, moderately alkaline (pH 8.2).
-------
Year
TABLE A-2. AVERAGE ANNUAL VALENCIA ORANGE YIELD (kg/tree) BY
REPLICATION FOR THE MINIMUM LEACHING EXPERIMENT AND THE
BORDER- AND CHECK-PLOT TREES IN SOUTHWESTERN ARIZONA
Treatment Replication
Treatment
average
5% Leaching
1974 124 124 132 138 105 110
1975 124 135 130 154 149 149
1976 63 45 70 68 81 86
108 105 98 116
152 149 143 143
70 97 83 73
10% Leaching
1974 132 139 137 122 129 125 124 117 108 126
1975 106 131 160 150 165 149 165 159 141 148
1976 48 62 61 57 69 79 68 81 107 70
20% Leaching
1974 143 138 123 117 106 112 122 119 117 122
1975 128 170 155 132 134 152 166 138 147 147
1976 75 58 60 59 68 62 88 80 90 71
Bubbler Border
West
row
1974 127
1975 140
1976 76
East Aver-
row age
118
142
76
122
141
76
Drip Border
West East Aver-
row row age
108 116
160 170
84 91
112
165
88
Flood Check
West Center East Aver-
row row row age
Fertilizer Check
West Center East Aver-
row row row age
1974
1975 116
1976 88
131
103
115
103
121
98
171
100
186
112
177
104
178
105
-91-
-------
TABLE A-3. VALENCIA ORANGE LEAF ANALYSIS SUMMARY FOR SEPTEMBER 1974
Element (cone.)
N %
P mmole/100 g
K meq/100 g
Ca " "
Mg " "
Na " "
Cl " "
S " "
B M-g/g
Fe
Mn
Zn
Cu
Optimum
Range
2.2-2.7
4-6
25-45
150-250
17-50
2.5-6.5
<4.2
12-19
25-150
60-150
20-100
25-100
4-10
Leaching Treatments
5%
2.61
3.95
25.1
236
19.3
1.8
<2.5
19.3
194
61
26
25
11
10%
2.77
3.82
24.4
241
19.3
1.9
<2.5
19.2
199
56
27
26
11
20%
2.77
4.02
25.8
237
19.4
1.9
<2.5
20.0
203
58
27
25
11
Border
Bubbler
2.64
3.82
26.2
243
19.7
1.8
<2.5
20.7
199
65
33
22
10
Trees
Drip
2.64
4.12
25.9
245
17.4
1.9
<2.5
19.5
179
66
31
20
12
Fertilizer
check plot
2.44
3.85
29.8
246
20.4
0.9
<2.5
20.8
186
58
21
21
10
Evalu-
ation
OK
OK
OK
OK
OK
OK
OK
OK
HIGH
OK
OK
OK
OK
-------
TABLE A-4. ANNUAL MEASUREMENTS OF VALENCIA OKANGE TREE TRUNK
CIRCUMFERENCE (mm) AVERAGED BY REPLICATION FOR THE MINI-
MUM LEACHING EXPERIMENT IN SOUTHWESTERN ARIZONA
Year
Treatment Replication
Treatment
average
5% Leaching
1973 551 574 564 566 589 582
1974 544 579 582 584 610 589
1975 576 614 610 616 635 621
1976 630 651 661 651 665 652
10% Leaching
1973 556 564 574 577 559 577
1974 582 579 577 599 566 599
1975 622 604 612 633 599 632
1976 651 653 652 666 635 670
20% Leaching
1973 572 566 582 538 561 574
1974 587 579 589 554 574 582
1975 620 616 630 590 610 613
1976 653 658 666 631 646 658
587 579
587 592
646 623
671 655
566
579
608
642
587 589 556
599 605 572
631 637 604
668 670 637
574 556 556
599 574 584
639 604 614
673 638 653
573
583
617
653
571
586
619
656
564
580
615
653
Bubbler Border
West
row
1973 561
1974 574
1975 614
1976 638
East Aver-
row age
561
574
614
640
561
574
614
639
Drip Border
West East Aver-
row row age
564 592 578
594 607 600
623 633 628
660 673 666
Flood Check
Fertilizer Check
West
row
1973
1974 536
1975 578
1976 596
East Aver-
row age
513 524
556 567
582 589
West
row
550
604
618
Center East Aver-
row row age
544 547
606
631
595
628
602
626
-93-
-------
Radial Distance from Tree, m
a.
a>
a
o
CO
North Radial
East Radial
Diagonal Radial __.
Figure A-l. Distribution of total head under the center citrus tree
of plot H4 on January 22, 1976.
-94-
-------
Q.
0>
Q
O
tn
Radial Distance from Tree, m
2
North Radial
Diagonal Radial
East Radial
Figure A-2. Distribution of total head under the center citrus tree
of plot H4 on March 23, 1976.
-95-
-------
Radial Distance from Tree, m
a.
a>
o
North Radial
Diagonal Radial
Figure A-3. Distribution of total head under the center citrus tree
of plot H4 on April 28, 1976.
-96-
-------
Radial Distance from Tree, m
North Radial
East Radial
Diagonal Radial
Flgure A-4. Distribution of total head under the center citrus tree
g of plot H4 on August 24, 1976.
-97-
-------
Q.
a>
Q
o
V)
.30
Radial Distance from Tree, m
I 2
North Radial
Diagonal Radial
East Radial
Figure A-5. Distribution of total head under the center citrus tree
of plot H4 in the morning on August 28, 1976.
-98-
-------
.c
"5.
a
o
c/j
Radial Distance from Tree , m
I 2 i
North Radial
Diagonal Radial
East Radial
Figure A-6. Distribution of total head under the center citrus tree
of plot H4 in the evening on August 29, 1976.
-99-
-------
Radial Distance from Tree, m
.30
.60-
.90-
1.20-
1.50-
1.80
.-ZOkPa
.30
Diagonal Radial
.60-
.90-
1.20-
1.50-
1.80
ik i M A L.
IN Or in
£
"5.
Q
"5
V)
.^w
.60-
.90-
1.20-
1.50-
i fln -
^^^^^
East Radial
Figure A-7. Distribution of total head under the center citrus tree
of plot H4 in the afternoon on September 1, 1976.
-100-
-------
Q.
0>
Q
O
C/J
Radial Distance from Tree, m
2
North Radial
Diagonal Radial
Figure A-8. Distribution of total head under the center citrus tree
of plot H4 on September 4, 1976.
-101-
-------
Q.
V
Q
Radial Distance from Tree , m
I 2
North Radial
Diagonal Radial
Figure A-9. Distribution of total head under the center citrus tree
of plot H4 on September 9, 1976.
-102-
-------
Distance from Tree , m
FEB '75 MAR '76
3.6 3.0 2.4 1.8 1.2 0.6 0 3.6 3.0 2.4 1.8 1.2 0.6 0
drip area , ,drip area
m«q/l 0-15 15-30 30-60 60-90 >90 PlotL7 Center Tree LF = .05
x = Salinity Sensors Locations
Figure A-10. Cross section of soil chloride distribution under the center tree
of a 5% (L7) leaching plot after 1 and 2 years.
-------
o
-p-
Distance from Tree , m
FEB '75
3.6 3.0 2.4 1.8 1.2 0.6 0
x-drip area
MAR '76
3.6 3.0 2.4 1.8 1.2 0.6 0
drip area
015
M Soil Chloride Distribution
3?60 60^90 >90 Pl°* H<* Center Tree LF=.20
* - Salinity Sensor Locations
Figure A-ll. Cross section of soil chloride distribution under the center tree
of a 20% (H4) leaching plot after 1 and 2 years.
-------
TABLE B-l. DESCRIPTION OF SOIL AT THE ALFALFA EXPERIMENTAL SITE
PROVIDED BY THE SOIL CONSERVATION SERVICE
Date: September 10, 1974. Area; Yuma County, Arizona.
Description by; E. Chamberlin and J. White.
Classification; Typic Torrifluvent, coarse-silty, mixed, calcareous, hyperthermic.
Location; 800'E, 900'S. of NW corner section 12, T. 8.S., R.16.W. of G. & S.R. B. L. & M.
Vegetation; Fallow. Climate; 2-4 inches precipitation, mean annual air temp. 70-75°F. Parent Mate-
rial; Gila River alluvium. Topography; Level. Elevation; 288*. Drainage: Well-drained.
Additional Notes; Indio very fine sandy loam (color for dry soil unless otherwise noted)
HORIZON DESCRIPTION
Ap 0-13 inches (0-33 cm); light brown (7.SYR 6/4) very fine sandy loam, brown (7.SYR 4/4) moist;
weak granular structure; soft, very friable, nonsticky, nonplastic; common medium and fine
roots; common medium tubular and many fine interstitial roots; slightly effervescent; moderately
i alkaline (pH 8.2); abrupt smooth boundary.
o
V Cl 13-21 inches (33-53 cm) light brown (7.SYR 6/4) silt loam, brown (7.5YR 4/4) moist; weak very
fine platy structure; slightly hard, very friable, slightly sticky, slightly plastic, few
medium and many fine roots; common tubular and many very fine interstitial pores; slightly
effervescent; moderately alkaline (pH 8.2); abrupt smooth boundary.
C2 21-33 inches (53-84 cm); light brown (7.SYR 6/4) silt loam, brown (7.SYR 4/4) moist; massive;
slightly hard, very friable, slightly sticky, slightly plastic; few medium and fine roots; few
tubular and many very fine interstitial pores; strongly effervescent; moderately alkaline (pH
8.5); clear smooth boundary.
C3 33-45 inches (84-114 cm); light brown (7.SYR 6/4) silty clay loam; brown (7.SYR 4/4) moist;
massive; hard, friable, sticky, slightly plastic; few tubular and many very fine interstitial
pores; strongly effervescent; moderately alkaline (pH 8.4); clear smooth boundary.
C4 45-72 inches (114-182 cm); light brown (7.SYR 6/4) silty clay loam, brown (7.SYR 4/4) moist;
massive; slightly hard, friable, sticky, plastic, few tubular and many very fine interstitial
pores; strongly effervescent; strongly alkaline (pH 8.8).
-------
M
O
I
CO
1.2
i.o H
0.8-
c
to* 0-6 H
o>
g 0.4-
rn 0.2
j I
I i I
I i
5% Leaching
0.9
A = Harvest Dates
A A A A A
A A A A A
\
80
I
160
1975
240
I '
320
I
40
I ' I '
120 200
1976
Julian Date
280 360
Figure B-l. Salinity trends with time for the 5% leaching treatment of the alfalfa experiment.
-------
)-
o
-J
I
CO
1.2
1.0 H
0.8 -
"o
CO 0.6 -
CD
0.4-
o
CO 0.2-1
\ I
I I
20% Leaching
A = Harvest Dates
A A A A A A
A A A A A
80
1
160
1975
240 320
I
40
I ' I I I
120 200 280
1976
JulIan Date
360
Figure B-2. Salinity trends with time for the 20% leaching treatment of the alfalfa experiment.
-------
o
oo
I
IRRIGATION, mm
- DRAINAGE, mm
LEACHING FRACTION
200
10
.02
SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL
1975 1976
Figure B-3. Accumulated values for irrigation, drainage, and leaching fraction
for the 5% alfalfa leaching treatment.
-------
I
l_l
o
1400
140
.28
1000
100
.20
600
60
.12
200
20
.04
I I I I I I I I
IRRIGATION,mm
DRAINAGE, mm
LEACHING FRACTION
I T
Ext I - 10 -
SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL
1975 1976
Figure B-4. Accumulated values for irrigation, drainage, and leaching fraction
for the 10% alfalfa leaching treatment.
-------
o
1400
280
.28
1000
200
.20
600
120
.12
200
40
.04
I I
I I \ I
IRRIGATION,mm
- DRAINAGE, mm
LEACHING FRACTION
Lys - H- 20
TH- 20-S
I I I I I
SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL
1975 1976
Figure B-5. Accumulated values for irrigation, drainage, and leaching fraction
for the 20% alfalfa leaching treatment.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
EPA-600/2-77-154
3. RECIPIENT'S ACCESSION>NO.
4. TITLE AND SUBTITLE
MINIMIZING SALT IN RETURN FLOW THROUGH
IRRIGATION MANAGEMENT
5. REPORT DATE
July 1977 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTI
U.S. Salinity Laboratory Staff
8. PERFORMING ORGANIZATION REPORT NO
U.S. Salinity Lab. Publ. #613
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Salinity Laboratory
Agricultural Research Service, USDA
Post Office Box 672
Riverside, California 92502
10, PROGRAM ELEMENT NO.
1HB617
11. CONTRACT/GRANT NO.
EPA-IAG-D4-0370
12. SPONSORING AGENCY NAME AND ADDRESS
Robert S. Kerr Environmental Research Laboratory-Ada,
Office of Research and Development OK
U.S. Environmental Protection Agency
Ada, Oklahoma 74820
13. TYPE OF REPORT AND PERIOD COVERED
Interim 12/4/73 to 12/5/76
14. SPONSORING AGENCY CODE
EPA/600/15
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Two field experiments are being conducted in southwestern Arizona to investi'
gate the potential of reducing the salt load in irrigation return flow by decreased
leaching. Three leaching treatments of 5, 10, and 20%, replicated nine times for
citrus and five times for alfalfa, were established and compared with conventional
flood irrigation management. Results on citrus indicate that leaching percentages of
8, 11, and 22 were achieved, compared to 47% on the border flood check. The best
estimate of the annual evapotranspiration of citrus is 1400 mm. Reduced leaching has
not adversely affected fruit quantity or quality. If leaching were reduced to 20%,
the volume of drainage from the 3000 ha of citrus in the district would be decreased
43.7 x 106 m3/yr and the salt load would be cut by 45,500 Mg/yr. Leaching percentages
of about 3, 5, and 10 have been obtained in the alfalfa. The level-basin flood check
received the same amount of water as the high leaching treatment. Results indicate
that the sprinkled plots were underirrigated and that the annual evapotranspiration
for alfalfa is about 2000 mm. Yields from the sprinkled plots have been 16% less
than those on the flooded field because of underirrigation, weed problems, and soil
compaction. Even with reseeding and less frequent irrigation, it is unlikely that
substantial improvement is possible over the low leaching obtained on the level-basin
flood check.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Irrigation, salinity, leaching, water
quality, desalting, efficiency
Colorado River Basin,
salinity control,
water use, Gila Project,
Wellton-Mohawk Division,
02C
B. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
2" NO. OF PAGES
127
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
-111-
* OS. GOVERNMENT PRINTING OFFICE 1977- 757-056/6480
------- |