United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada, OK 74820
Bft
Research and Development
EPA-600/S2-82-073 Nov. 1982
Project Summary
Minimizing Salt in Return
Flow Through Irrigation
Management
Because irrigation is the largest user
of water, irrigation management can
have a significant impact on water
quality. This report describes a
research project designed to demon-
strate that, through improved
irrigation water management, it often
is possible to reduce the salt load in
irrigation return flow and, thereby, the
concentration of dissolved solids in
downstream reaches of a river system.
The study was undertaken with a
view to reducing the salinity of the
Colorado River and, more specifically
to address alternative means of
meeting the U.S. obligation to deliver
to Mexico Colorado River water of
specified quality and quantity. Other
concurrent studies are underway in
Arizona by the U.S. Bureau of
Reclamation, the Agricultural
Research Service and the University
of Arizona; an extensive program of
planning and technical assistance is
being conducted by the Soil
Conservation Service. These studies
complemented similar studies by
various entities in the Upper Colorado
Basin. Together, these studies
highlight the potential of enhancing
water resource use by stressing
measures to increase on-farm water
management.
Two field experiments were
conducted in the Wellton-Mohawk
Irrigation and Drainage District of
southwest 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 percent, replicated nine
times for citrus and five times for
alfalfa, were established and
compared with conventional flood
irrigation management.
The annual evapotranspiration of
mature Valencia orange trees was
found to be about 1470 mm. The
leaching fractions, determined
indirectly by several techniques, were
close to those intended. The leaching
fraction computed for the border flood
check was 0.47. The leaching
fractions in the alfalfa experiment
were not exactly as planned. Based on
several independent measurement
techniques, the actual leaching
fractions achieved were 0.06, 0.10,
and 0.14; near 0.11 for the flood
check. Annual evapotranspiration was
estimated to be 1930 mm. After
reducing the frequency and increasing
the amount of water applied with each
irrigation, no differences in alfalfa
yield were noted among treatments or
with the flood check.
Both the citrus and the alfalfa
experiments verified that Water
applications could be reduced
substantially below average
commercial practice without loss of
crop yield. Thus, irrigation return flow
and its inherent salt load can be
reduced consistent with the leaching
requirement for the crop and irrigation
water in question.
This Project Summary was devel-
oped by EPA's Robert S. Kerr Environ-
mental Research Laboratory, Ada,
OK. to announce key findings of the
research project that is fully docu-
mented in a separate report of the
same title (see Project Report ordering
information at back).
-------
Figure 1. Location of citrus and alfalfa minimum leaching experiments in southwestern Arizona.
Introduction
The salt load of the Colorado River
constitutes a major problem. Unless
corrective actions are taken, the current
average salt concentration of about 850
mg L-' in the lower reaches of the
river is expected to increase substa ntial-
ly as further water development takes
place. Such an increase would have
serious economic consequences for the
seven states adjoining the river and for
the Republic of Mexico. Return flow
from irrigated lands has been identified
by the U.S. Environmental Protection
Agency (1971) as a major contributor of
salts.
Research at the U.S. Salinity
Laboratory lead to the conclusion that
often the amount of salt in drainage
water from irrigated agriculture can be
reduced substantially by modifying
irrigation management. This project
was initiated to evaluate the pertinence
of this conclusion with particular
reference to the Wellton-Mohawk
Irrigation and Drainage District in
southwestern Arizona, a district where
improved irrigation management is a
potential alternative to construction of a
desalting complex to alleviate the
adverse effects of returning brackish
drainage water to the river.
The primary objective of these field
studies was to determine the feasibility
of reducing the salt load in drainage
water by reduced leaching while
maintaining crop yield. Achieving this
objective implied increasing irrigation
efficiency, using uniform and frequent
irrigations with minimal leaching
consistent with crop needs; and taking
advantage of the reductions in mineral
dissolution or increases in salt precipi-
tation as the leaching fraction is
reduced. Additional objectives were to
determine the components of the water
and salt balance under the improved
irrigation regimes and to determine the
requirements of an irrigation system for
these purposes.
One experiment was installed in
December 1973 with citrus on coarse-
textured mesa soil and continued
through fruit harvest in the spring 1979;
the second was started in September
1974 with alfalfa on medium-textured
valley soil and discontinued in October
1978. Concurrent research, by other
groups and by the U.S. Salinity
Laboratory, in the same district and in
Colorado, addressed corollary issues or
parallel questions under different cli-
matic and geological circumstances.
Thus, a concerted effort was mour
during the 1970s to address the rela
between agricultural water mana
ment and the salinity of the Color
River.
The final report describes the
experiments mentioned, reports
results and interprets the findings.
all objectives were fully attained, but
project succeeded in verifying the bŁ
premise. It also provided use
estimates of the potential changes
the water and salt balances that co
realistically be achieved.
Results and Conclusions
To investigate the potential
reducing the salt load in return fl
from irrigated areas by redut
leaching, two field projects w<
established near Tacna, Arizona, in 1
Wellton-Mohawk Irrigation and Dra
age District (Figure 1). Colorado Ri<
water with about 944 mg L"1 total d
solved solids (determined analytica
equivalent to 859 mg L"1 TDS by residi
was used for irrigation.
Citrus
In one project, trickle irrigation w
used to control the amount of wai
applied to each tree in a matt
-------
Valencia orange orchard located on
Dateland fine sandy loam soil. Three
treatments replicated nine times,
intended to result in leaching fractions
of 0.05, 0.10, and 0.20, were
established and compared with
conventional borderflood irrigation. The
trickle irrigated plots were irrigated
several times a day with a depth of 1.7
mm per irrigation. The irrigation regime
was imposed in December 1973 and
maintained until fruit harvest in April
1979.
The annual evapotranspiration of the
mature citrus trees was found to be
close to 1470 mm, a value somewhat
higher than indicated in earlier reports.
The actual leaching fractions obtained,
in contrast to those planned, could not
be measured directly. They were
estimated from soil salinity, soil water
chloride concentration, the salinity of
the percolate in buried extractors, and
from water balance data. A rather wide
range of estimates resulted, but they
were consistent with the conclusions
that the leaching fractions obtained,
averaged over time, were very nearly
those planned.
In the first four years, no effect of
irrigation treatment was observed on
fruit yield or quality. In the last year, a
low yield year for all treatments, the
yield of the five percent leaching
treatment was substantially lower than
that of the others. Unfortunately, an
unambiguous conclusion on the effect
of low leaching levels on citrus yield
would have required continuing the
experiment for additional years. The
data at hand suggest that the fifth year
yield depression at low leaching was a
real treatment effect. It is possible, but
not at all certain, that the capacity of the
tree trunks to store chloride delays the
negative effect on fruit yield several
years after the roots are exposed to
higher salinity levels.
With some uncertainty, we conclude
that the leaching requirement for
Valencia oranges irrigated with
Colorado River water exceeds five
percent and is less than 20 percent.
Foliar analysis indicated that nitrogen
applications, in the form of foliar urea
sprays of 690 g per tree per year,
somewhat exceeded the nutritional
requirement. Compared with University
of Arizona results in an adjacent
orchard that 340 g was borderline, a
reasonable conclusion would be that
0.5 kg per tree is an adequate level for
nitrogen fertilization.
The spatial heterogeneity, the
multidimensional nature of both the
root- and flow-systems, and the
temporal cycling prevented the
quantitative use of soil salinity data in
determining a clear partitioning of the
salt in depth and time. However, the data
in general were not at variance with
expectations. It can be concluded
unequivocally that insufficient time
elapsed to establish chemical
equilibrium in the soil solution with
respect to Na, Ca, and Mg, and thus the
sodium-adsorption-ratio.
A detailed study of water and salt
transport through the root zone was
made for one set of nine trees by
interruption of the high frequency
irrigation for extended periods. Of the
water uptake below the surface area
wet by the irrigation system (less than
half the total surface area), about 80
percent took place above a depth of 0.60
m. Significant uptake was noted outside
the wetted area, with water provided by
lateral flow. In summer, serious stress
developed in the trees when irrigation
was suspended more than 30 days; in
winter, interruption up to 60 days
showed no ill effects. These results
indicate that even in extremely hot
climates, citrus under frequent
irrigation is not particularly vulnerable
to temporary breakdowns of the
irrigation system. They also suggest
that, if water supplies are limited, it may
be practical to practice deficit irrigation
in the season of peak evaporative
demand, depending on water stored in
the soil profile that is replenished in the
cool season; adequate fruit yield data
were not obtained to verify this
suggestion.
Root length distribution data showed
a similar pattern to that obtained from
water, hydraulic head and salt
measurements, with substantially more
roots at shallow depths (<0.3 m) under
the wetted area, and with 75 percent of
the roots above 0.6 m. In the flood
irrigated area, root distribution was
more uniform with depth and also with
distance from the trunk.
Alfalfa
The second project dealt with
irrigation of alfalfa on Indio fine sandy
loam soil. About 2 ha of an 8-ha field
was divided into 15 plots, providing five
replications of three treatments. The
treatments imposed were expected to
yield leaching fractions of 0.05, 0.10,
and 0.20. Irrigations were applied with a
moving-boom spray system that
originally applied a depth of 6 mm of
water per pass; after modification, it
applied 70 mm per pass. The remainder
of the field was irrigated by level basin
flooding, applying about 150 mm per
irrigation. Alfalfa was seeded early in
October 1974 and replanted in October
1976. The experiment ended in October
1978.
The original high frequency irrigation
schedule proved incompatible with the
other cultural practices. It resulted in
excessive compaction and weediness
and consequently, in reduced crop
yields and poor infiltration. After the
irrigation system was modified to
accommodate less frequent irrigation,
useful results were obtained.
As with the citrus, evapotranspira-
tion was estimated by a number of
computational procedures. The best
estimate is 1930 mm yr1. The leaching
fractions actually obtained were not
exactly as planned. A reasonable
estimate of the actual leaching fractions
is 0.06, 0.10, and 0.14; for the flood
check, it probably was near 0.11.
Drainage from the nearby mesa
caused relatively high water tables in
the alfalfa field part of the time.
Hydraulic head data showed that, in the
0.05- and 0.10-leaching fraction
treatments, the plants used some water
from the water table. No such evidence
was found for the 0.20-leaching
fraction plots. After replanting, no dif-
ferences in alfalfa yield were noted
among treatments or between the
sprinkler irrigated and the flood
irrigated alfalfa. The annual yield was
about 22 Mg ha-1. Thus, the leaching
requirement, using Colorado River
water, is less than five percent.
Soil samples showed that the field
had been well leached prior to the
experiment. After establishing the new
regimes, the salinity near the surface
face was lower in the flood check than in
the sprinkled plots. Soil water
chloride data indicated that, in the
sprinkled plots, 90 percent of the water
was taken up above a depth of 0.6 m.
Only 66 percent of the uptake took place
above that depth for the flood check.
Root density measurements gave a very
similar pattern for the sprinkled plots,
but we could not distinguish a different
pattern for the flood check; this could
have been due to the small number of
root samples analyzed.
The management of the farmer-
irrigated flood check with a leaching
-------
fraction of slightly over 0.10 illustrates
clearly that high irrigation efficiencies
can be obtained and are being obtained
commercially. In fact, no advantage can
be claimed for the sprinkler system for
use on alfalfa. The average efficiency for
alfalfa estimated by action agency
personnel, however, is substantially
lower than the 89 percent found here.
Thus, even if the controlled leaching
plots did not use significantly less water
than the flood check, the conclusion still
is that, especially with the locally well
adapted level basin irrigation system,
there is ample opportunity for reductions
in water use in the Wellton-Mohawk
District.
Both the citrus and the alfalfa experi-
ments verified that water applications
could be reduced substantially below
average commercial practice without
crop yield reductions. Presumably, the
same conclusion could be applied to
other crops grown in the area. The
consequences of such a reduction, if
widely effected, would be several.
Reduced water applications would
result in reduced deep percolation and
thus, in a reduced drainage flow. In the
short run, this reduced flow would have
the same salt concentration as at
present. At steady state, the reduced
flow would have a higher salt
concentration than the current drainage
flow, because of the salt stored in the
aquifer. The concentration would be
substantially higher than that expected
ultimately were the current water bal-
ance maintained.
Given a set of assumptions,
calculations can be made of the
changes in salt and water balances per
ha for each crop. Suffice it to point out
that reducing the drainage flow
consistent with the leaching
requirement reduces the volume of
water required to remove the necessary
salt from the system and thus reduces
the cost of disposal. By the same
amount that the drainage volume is
reduced, the amount of water retained
in the river for later use is increased.
We propose that the results obtained
in these experiments provide evidence
to support the assertion that Colorado
River diversions into the Wellton-
Mohawk District and, consequently,
average drainage flow, can be reduced
sufficiently, without damaging
agricultural production, to make
questionable the efficiency of alterna-
tive means of salvaging the drainage
flows.
This Project Summary was authored by the U.S. Salinity Laboratory Staff of the
U.S. Department of Agriculture, Riverside, CA 92501.
James P. Law, Jr., is the EPA Project Officer (see below).
The complete report, entitled "Minimizing Salt in Return Flow Through Irrigation
Management," (Order No. PB 82-257 445; Cost: $ 16.50, subject to change)
will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
U.S. GOVERNMENT PRINTING OFFICE- 1982 659-017/0862
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS 0000329
U S ENV1R PROTECTION A6ENC*
REGION 5 LIHRARY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |