Practical Applications of Drip Irrigation

In this chapter we will give a critical overview of the applications of drip irrigation
to the main branches of agriculture and horticulture. It is our aim to sum up the expe-
rience attained by this irrigation method since its evolution and development, and to
look at the problems encountered with specific crops under varying circumstances.

6.1
Orchards and Ornamental Trees

Drip irrigation is particularly suitable for tree crops. Trees are usually plan ted in rows
spaced widely enough to allow cultural operations such as spraying, pruning, picking,
etc. With traditional irrigation systems (gravity and sprinkler) the whole orchard,
including the inter-row spaces, is wetted, thereby interfering with management oper-
ations, damaging soil structure, and resulting in soil compaction, particularly if traffic
occurs soon after wetting. Moreover, problems of soil aeration due to excess water
may occur because the whole soil surface is wetted. These problems do not exist in
drip irrigation since here adequate aeration is provided at the fringes of the wetted
zone (Bravdo and Proebsting 1993). Moreover, water is applied only in the tree rows,
leaving a dry strip between rows. If there is no rain during the irrigation season, very
little weed growth takes place in the inter-row spaces. Fertilizers can be applied in a
very precise manner through the drip system. Drip irrigation is also well-adapted to
provide supplemental irrigation in humid areas because of the permanent installa-
tion, the relatively low pressure requirements and the ease of operation. Bucks (1995)
estimated, based on available statistics, that more than 10% of the main tree crops
worldwide are irrigated by microirrigation, which is a much higher percentage than
the fraction of the total irrigated area (see Table 1.1).

6.1.1
Citrus

The response of citrus to irrigation has been investigated, both under Mediterranean-
type climates, with dry summers and rainy winters, and under more tropical condi-
tions, with summer rains. Several authors have reported the evapotranspiration of
citrus based on water balance measurements. Van Bavel et al. (1967) estimated the
cumulative water use of mature orange orchards in Arizona to be 66% of the annual
Class A pan evaporation. Kalma and Stanhill (1972) found this to be 54% in the coastal
plain of Israel. Irrigation experiments summarized by Shalhevet et al. (1981) gave
similar results.

S. Dasberg et al., Drip Irrigation

© Springer-Verlag Berlin Heidelberg 1999
126 Practical Applications of Drip Irrigation

The tradition al methods of citrus irrigation are surface (basin and furrow) over-
head sprinkler. More recently, below-canopy sprinkler, minisprinkler, microjet and
drip systems have been introduced. The question arose as to the extent that partial
wetting provided by microirrigation can satisfy the transpirational demand of the
trees and reduce evaporation from the soil? Mantell (1977) reviewed the earlier work
on partial wetting of the soil surface in Israel. He found that alternate row irrigation
provided similar results to those obtained by irrigating every row, using the same
amount of water with a doubled irrigation interval. Dasberg (1995) compared several
long-term experiments carried out in farmers' groves. The results are shown in Table
6.1. In all these experiments, sprinkling was applied by low-angle sprinklers below the
canopy, wetting 90% of the soil surface, spray was applied by one sprayer per tree and
drip by 41!h emitters 75-100cm apart. The irrigation frequency was adapted to the
wetted volume as given by the irrigation method, with the same amount of water
applied for all irrigation methods. These data show that in no case did partial wetting
of the surface decrease or increase yields on a long-term basis. The response to applied
water was similar, which means that no water saving was achieved by partial wetting;
the trees used the same amounts of water in all cases.
Experience has shown that the conversion of the irrigation system of established
orchards from complete to partial wetting of the soil surface by drip or minisprin-
klers (30-40% wetted areal has no adverse effects (Bielorai 1982; Bielorai et al. 1985;
Bravdo and Proebsting 1993); the citrus root system, even of mature trees, adapts itself
quickly to the smaller wetted volume.
In an irrigation experiment with young Valencia oranges in Arizona, the trees irri-
gated by drip and basin methods showed a larger growth rate during the first 5 years
than the trees irrigated by sprinkler and f100d (Rodney et al. 1977). Subsequent yields
were significantly greater for trees irrigated by the first two, rather than the last two
methods. The drip system was operated daily, while the irrigation interval with the
other methods was 1 to 2 weeks. In another experiment with young grapefruit trees,
conducted for 9 years in Israel (Leitman et al. 1979), no appreciable differences in tree

Table 6.1. Long-term comparison of irrigation methods and quantities on fruit yields
(Dasberg, 1995)

Experiment Years Applied Citrus yield (Mg/ha)

Water (mm) Drip,one Drip, two Spray Sprinkler

lateral laterals

Grapefruit 4 630 87 91 83
(Yagev, 1977) 800 99 100 94
Grapefruit 7 521 66 68 69
(Bielorai 1977) 707 80 77
895 82 83
(Bielorai et al. 1985) 7 575 64 60
765 71 77
900 82 76
Oranges 610 58 57 53
(Raber et al, 1990)
Mineola 6 622 53 64 55
(Dasberg et al. 1994) 504 53 60 57
Orchards and Ornamental Trees 127

development and yield were observed between drip, sprinkler and minisprinkler,
using the same irrigation schedule.
The yield response of a mature grapefruit orchard to sprinkler and drip irrigation
with one (Drip I) or two (Drip II) laterals per tree row on a sandy loam soil are shown
in Fig. 6.1 (Bielorai 1977). He found that yield, fruit quality and tree development were
similar in drip (with wetting of only 30% (one lateral) or 40% (two laterals) of the soil
surface) and sprinkler irrigation, provided the irrigation is carried out at short inter-
vals (2-3 days).At greater intervals (7 days) drip irrigation with one lateral gave some-
what lower yields. Similar results were obtained by Yagev (1977).
A comparison of drip, microjet, sprinkler and basin irrigation was made on mature
Navelorange trees in South Africa. Results showed similar yields for daily drip irri-
gation compared with microjet and sprinkler with two-week application intervals
(Fouche et al. 1979). Results of an experiment with mature Navel oranges on sandy
day loam soil in California showed no difference in fruit production and quality
between drip, furrow and sprinkler irrigation (Aljibury 1981). With drip irrigation,
the soil water potental was kept at -10 kPa, while in the furrow and sprinkler plots
irrigation was applied when the soil water potential at 30 cm depth reached -70 kPa.
Nevertheless, the effective amount of water applied by drip irrigation was smaller than
by either sprinkler or furrow.
Not all attempts to introduce drip irrigation in mature orange groves have been
successful. Cole and Till (1977) reported unsatisfactory results with one drip lateral

~ ::t ~~~ ~DRIP

SPRINKLER

9 11

~~.2
> 90 E § 1 9 7 3
~

80 ~19141
1915
, I

100

~~350l·~ ~ ~1
~2'300-~4 ~
~
I/l 250
I,2 I
3
I
7
L-J--1
2 3 1 7 \1, 21
IRRIGATION FREQUENCV (days)

Fig.6.1. Yield indices of drip and sprinkle irrigated grapefruits at Sa'ad, 1973-1977 (Bielorai
1977)
128 Practical Applications of Drip Irrigation

per row on coarse sand in Australia because only 25% of the area was wetted. The 60%
wetting necessary to achieve good production could be attained only by two laterals
per row (Cole and TillI977). In Florida, citrus is planted on sandy soil with low water-
holding capacity. Supplementary irrigation is beneficial despite an annual pre-
cipitation of 1250-1400mm. Supplemental irrigation in high-density orange groves
(spacing (2.4 x 3.4m) in Florida on sandy soil was most effective when 70-80% of the
area under the tree canopy was irrigated (Koo 1978). In a low-density grove ofValen-
cia oranges, spray irrigation covering 30-50% of the surface area gave higher pro-
duction than drip irrigation, which covered only 5-10% of the area in a 5-year
experiment. Both treatments yielded more than the control without irrigation,
although the average precipitation was 1285mm (Koo and Smajstrla 1985). These
studies show the importance of covering sufficient soil area by drip irrigation in a
humid area with sandy soil, where irrigation is supplemental. The roots proliferate in
the whole soil volume, but drip irrigation wets only part of the root system. In Florida,
microsprinklers are often preferred over drip systems, not only because they provide
greater soil coverage, but also because they can provide some degree of freeze pro-
tection (Boman and Parsons 1995; Smajstrla 1993).
The above results indicate that citrus roots adapt themselves quickly to different
patterns of water application, provided the overall transpirational demand of the tree
is met. Litde information is available on the direct transpiration of citrus trees; there-
fore, estimating what fraction of the total evapotranspiration results from direct evap-
oration from the soil is difficult. The available data do not indicate any water saving
in orchards by drip irrigation compared with other methods. This implies that, with
drip, more frequent water application is necessary than in the traditional methods
because of the more limited root volume. Drip appears to be a very good practice for
young orange trees. However, with mature trees, adequate soil coverage should be pro-
vided, especially when drip irrigation is supplementary. One lateral per tree row is
sufficient in most cases, provided fiow rates and spacing are adjusted to the hydraulic
properties of the soil, as explained in section 4.2. Aljibury (1981) estimated the cost
of drip compared with furrow and sprinkler irrigation for citrus in California. The
drip and furrow methods used virtually the same small amounts of energy per hectare
for direct pumping, whereas the sprinkler system was the highest energy user. A com-
parison of drip, under-tree spray and overhead sprinkler for supplemental irrigation
in Florida (Myers 1977) showed initial and operating costs of a drip system to be lower.
Drip irrigation had a higher irrigation efficiency, lower water use and lower operat-
ing pressure, resulting in a lower power requirement (see Table 6.2).

Table 6.2. A comparison of irrigation systems for supplemental irrigation of orchard craps in
Florida (Myers 1977)

Type of system Drip Under tree spray Overhead sprinkler

Installation costs/ha 865$ 1235$ 2965$

Water supply costs/ha 250$ 450$ 531$
Operating pressure (kPa) 83 104 380
Irrigation efficiency (%) 95 85 70
Labor (man h/ha year) 15 5 4
Orchards and Ornamental Trees 129

Irrigation generally causes an increase in fruit number and size. Excess irrigation,
on the other hand, causes excessive vegetative growth and production of small fruit
high in juice and of low keeping quality. Medium water stress (deficit irrigation) does
not cause any yield reduction, but tree growth is impaired. Severe stress causes lower
fruit set and lower yields (Hilgeman 1977; Castel and Buj 1990; Dasberg and Erner
1995).

6.1.2
Avocado

Avocado culture is developing rapidly in South America, Mexico, Florida, California,

South Africa, Israel, North Africa and Spain. One of the first drip irrigation
experiments, carried out in an avocado orchard in San Diego County (Gustafson
et al. 1980), compared drip with sprinkler irrigation in a commercial orchard on
sloping land. They used one sprinkler per tree with water applied weekly, or three
emitters (with dis charge of 4l!h) per tree with water applied daily. The water appli-
cation was regulated by tensiometers at 30 and 60 cm depth. The drip system
performed satisfactorily on the sloping land, using appreciably less water in the
early years. Tree growth was similar for the two methods. However, fruit yield
was somewhat higher for the sprinkler-irrigated trees, probably because the soil
volume wetted under drip was too small. Another experiment comparing drip
with sprinkling for avocado on a heavy soil was carried out in Israel (Levinson
and Adato 1991). They found the dry drip treatment, irrigated intermittently
during daylight (0.46 Ep,n), to be superior to the wet drip treatment, irrigated
similarly with larger amounts of water (0.64 Ep,n), and to the sprinkler treatments
irrigated every 4 days. They related these differences to better root development
and better soil aeration (Levinson and Adato 1991). An experiment comparing
three levels of irrigation with drip (0.3, 0.6 and 0.9 Ep,n) was carried out in Crete.
The low irrigation treatment showed lower root density, insufficient wetted
volume and lower yields than the other treatments; the high irrigation treatment
showed excessive deep percolation and similar yields as the 0.6 Er,m treatment
(Michelakis et al. 1993).
A long-term experiment, on a day soil in the Western Galilee on trees planted
in 1974, was begun in 1978 with four amounts of water applied daily - 60, 80,
100 and 120% replenishment of soil water depletion from the 60-cm soil depth.
Tree growth and yields were high er as the water application was increased during
the last 5 years of the experiment when the trees had reached full bearing.
The main condusion was that, even with the highly efficient daily drip irrigation,
no water saving could be achieved (Lahav and Kalmar 1991).
Another experiment was carried out on trees plan ted on a sandy loam loessial soil
in the northern Negev area. Microjets (one per tree) were compared with drip irriga-
tion (eight emitters with a discharge of 4l!h per tree) using different amounts of water.
The results of the 10-year experiment showed greater tree growth and high er yields
with larger amounts of water applied by drip compared with microjet irrigation. Drip
irrigation also caused less leaf tip burn and chlorosis than microjet irrigation (Tomer
et al. 1995).
130 Practical Applications of Drip Irrigation

6.1.3
Deciduous Trees

Deeiduous trees are grown in both temperate and semiarid eonditions. An experiment
with apples, eomparing daily drip irrigation with daily and bi-weekly sprinkler
irrigation was earried out at Prosser, Washington, where the summers are dry
(Middleton et al. 1979). The amount of water applied to the drip-irrigated trees was
determined by the water use of trees grown in lysimeters, while the amount applied
by sprinkler was determined by daily pan evaporation. The results showed earlier
tlowering under drip (in the third and fourth years) than under sprinkler irrigation
(see Table 6.3). Consequently, during the first 6 years of growth, yields were mueh
higher under drip than under sprinkler: trees used less water, developed better shapes
and required less pruning and spreading. Moreover, drip-irrigated trees yielded fruit
of high er quality (both for sale as fresh produee and for preserving) than sprinkler
irrigated trees (Drake et al. 1981). This is one of the few well-documented eases
showing a specifie advantage of drip versus sprinkler under the same irrigation fre-
queney. The authors attributed the differenees to ehanges in root proliferation. The
drip-irrigated trees were supplied with one emitter per tree, having a diseharge of 4
lIh, whieh limited the root volume per tree to 0.7m 3 , as compared to a maximum
possible root volume of 13 m 1 per tree (Proebsting et al. 1977).
The response of young almond trees to different initial levels of drip irrigation
(ranging from 209 to 562 mm) was evaluated in Spain. Trunk growth, eanopy area and
yield were alilinearly related to the amount of irrigation water applied (Torreeilas et
al. 1989). Another drip irrigation experiment on young almond trees was earried out
in California, with a larger range of water applieations (269 to 913 mm annually, during
the fifth through seventh year after planting). Trunk growth was a linear funetion of
the total amount of water applied during the experiment. However, yields increased
only up to an intermediate level of water applieation (790 mm, Hutmaeher et al. 1994).
These data show that, in almonds, with their short growing season, fruit yield is closely
related to eanopy development.
Drip is suitable for supplemental irrigation in areas with summer rainfall. In dry
seasons, drip-irrigated peaeh trees gave higher yields than sprinkler-irrigated (onee,
before harvest) trees and non-irrigated eontrols. The praetiee of drip irrigation
aeeording to evaporation demand also resulted in a greater inerease in trunk diame-
ter (Reeder et al. 1979). Similar results were obtained with peaeh trees in Georgia
(Chesness and Couvillon 1980). These authors used four emitters per tree with

Table 6.3. Water use, growth and yields of Redspur Delieious apples under three water regimes
(Middleton et al. 1979; Evans and Proebsting 1985)

Irrigation rnethod Drip, daily Sprinkler, daily Sprinkler, bi-weekly

Water applieation 1973-1976 (ern) 34 304 220

Water applieation 1977-1978 (ern) 53 159 150
Trunk area inerease 1976 (ern') 14.8 17.7 16.5
Fruit yield 1976 (kg/tree) 8.8 1.0 0.4
Fruit yield 1978 (kg/tree) 45.7 19.5 8.4
Average yield 1978-1984 (kg/tree) 51.3 38.2
Field and Fodder Crops 131

Q = 4l!h. They showed the importance of injecting nitrogen into the irrigation water,
thus saving appreciable amounts of fertilizer.
Drip installation for supplemental irrigation in cases of drought or deficient rain-
fall was reported to have definite economic advantages over overhead sprinklers (Funt
et al. 1980). Installation costs, pumping costs and water requirements were alliess than
half of those for sprinklers. Moreover, larger areas can be covered by drip, compared
with sprinklers with the same water ftow rate. Kipp (1992) reviewed the development
of irrigation and fertilization in apple orchards in the Netherlands. Drip irrigation
seemed to be an excellent method for supplementary irrigation in dry years. Fertiga-
tion resulted in strong shoot growth and larger production because it optimized the
nutrient availability in the root zone (Kipp 1992).

6.1.4
Vineyards

Drip irrigation is particularly suitable for grapevines, since the water can be applied
to the vine rows without interfering with cultural operations and without wetting the
fruit and foliage. An experiment on young vines comparing daily drip with sprinkler
and ftood irrigation in California (Peacock et al. 1977) showed no difference between
the development of the vines as measured by trunk circumferences with all three
methods. Water use was much less with drip irrigation, however. During the first year
of fruit bearing, no differences occurred in yield and quality between the three
methods. Another experiment with older grapevines was carried out in British
Columbia (Stevenson 1981). The transition from weekly sprinkler to daily drip irri-
gation did not cause any changes in growth or yield. It was shown that nitrogen can
be applied through the drip system and that the water use efficiency is improved by
this irrigation method.
An extensive investigation comparing several drip irrigation modes (three fre-
quencies, one or two emitters per vine and three water quantities) with conventional
furrow irrigation with the same three quantities of water was carried out in Arizona
(Bucks et al. 1985). Some of the results are given in Table 6.4. These data show that
yields and fruit quality were slightly increased with two emitters per vine over furrow
irrigation, whereas yield with drip irrigation with one emitter per vine did not differ
from furrow irrigation. A trend of increased yield for 3- or 6-day frequencies over
daily drip irrigation occurred in 3 out of 4 years. Plant water stress developed and
yield was decreased with water applications at 25% less than the seasonal water appli-
cation (Bucks et al. 1985). It can be concluded that drip irrigation is a suitable method
for the growth of vines for grapes as well as for wine production. A large percentage
of vines worldwide are drip irrigated.

6.2
Field and Fodder (rops

6.2.1
Cotton

Cotton is a major irrigated crop grown in tropical, sub-tropical and arid regions. Most
of the irrigated cotton areas in the world use gravity methods, although in the USA
132 Practical Applications of Drip Irrigation

Table 6.4. Comparison of grape yield for drip and furrow irrigation (Bucks et al. 1985)

Comparison 1973 1974 1975 1976

Two emitters vs. one +9% +9% + 17% NA

emi tter per vine
Two emitters vs. + 13% +7% + 18% NA
furrow irrigation
One emitter vs. NS NS NS NA
furrow irrigation
Different drip NS +12% for + 15% for 1.25 + 23% for 1.25
irrigation quantities 1.35 & 1.05 and 1.0 over and 1.0 over
over 0.8 ET 0.75 ET 1.5 ET
Different furrow NS Trend for 1.35 +29% for 1.25 NS
irrigation quantities and 1.05 & 1.0 over
over 0.8 ET 0.75 ET
Different drip Trend for 3 Trend for 3 NS + 13% for 6 days
frequencies and 6 days and 6 days over daily
over daily over daily

NA = no comparison made, NS = no significant differences, ET = seasonal evapotranspiration

and Israel, sprinkler irrigation is used extensively. In recent years, however, drip has
been introduced in Israel and has replaced sprinkling to a large extent, as will be
shown below. Wierenga (1977) compared surface and drip irrigation in New Mexico
and found that drip irrigation resulted in 8% higher cotton yields than did surface
irrigation, using 24% less water at 100% efficiency. Subsequently, drip irrigation was
compared with the conventionallevel-basin method in Arizona (French et al. 1985).
It was found that both methods, when properly managed, have a potential for high
cotton yields and water use efficiency. Laterals every second row (2 m apart) were sat-
isfactory, but one lateral for every third cotton row was inadequate on these course-
textured soils (French et al. 1985). Frequent drip irrigation during fruiting enhanced
yields because the period of flowering was lengthened. This could also be achieved by
doubling the number of flood irrigations (Radin et al. 1992). Subsurface drip was
developed on a commercial scale in Arizona (Tollefson 1985). Drip-tapes were buried
at 25 cm depth under each cotton row. This reduced salt concentration near the plant
rows, and enabled crop rotation, fertilizer injection and computer contro!, resulting
in appreciable water savings. Drip-irrigated fields out-yielded furrow irrigation by a
30% average over four years (Tollefson 1985). Subsurface drip irrigation of cotton was
also developed in Texas, resulting in increased yields over surface irrigation and
decreased labor, cultivation and herbicide costs (Hengeller, 1995). An additional
benefit of drip irrigation is the possibility of growing cotton by frequent irrigation
with highly saline water (Ayars et al. 1985). Good results were also obtained by drip
irrigating cotton on saline soils in Spain (Fereres et al. 1985). In the southeastern
coastal plain of the USA, cotton irrigation is supplementary. An irrigation system for
humid areas should have a low labor requirement, low capital investment and the
capacity to sustain the crop during drought at critical periods. Subsurface drip irri-
gation, using the same laterals for many years and wide lateral spacing, can answer
these requirements, as has been shown by several field experiments (Camp et al.
1995b).
Field and Fodder Crops 133

Table 6.5. Survey of drip and sprinkle irrigation of cotton in two regions in Israel (Cohen
1978-1986)

Region Year Irrigated area (ha) Seed cotton yield (kg/ha) % Difference
Sprinkle Drip Sprinkle Drip

Lachish 1981 4000 500 4540 5120 13

1982 3075 llOO 4820 5320 10
1983 2900 1750 4990 5270 7
1984 800 1600 4400 4960 13
Gallilee 1983 7960 1900 5180 5580 8
1984 3410 2400 4200 5380 28
1985 2570 3100 4690 5550 18
1986 l330 3600 4960 5410 9

Israel, although producing only a sm all fraction of the world cotton crop, produces
record yields (see Table 6.5). Cotton is the main irrigated crop in Israel, occupying a
large portion of irrigated areas. Until recently, most of the cotton was irrigated by
sprinkler, using tow-moved lines, with the water requirements and scheduling deter-
mined by extensive experimentation (Shalhevet et a!. 1981). During the past decade,
drip has replaced sprinkler irrigation on most of the fields. The annual reports of the
Cotton Research Board show the development of drip irrigation in cotton (Cohen
1978-1986). Originally, drip was applied mostly on shallow soils with low water-
holding capacity, stony soils, coarse-textured soils, fields of irregular size or on steep
slopes which were difficult to irrigate with the tractor-towed sprinkler lines. Field
experiments showed that drip irrigation resulted in somewhat higher yields than
sprinkler with the same amounts of water, usually about 10% more in seed cotton.
This comparison is valid only under the most favorable conditions for sprinkler, i.e.,
irrigation without wind (preferably at night) and with adequate pressure. The drip
system for cotton is usually designed with emitters having Q = 41!h, spaced 1 m apart
along the row, one lateral for every two rows; while the sprinklers are generally spaced
18 x 12m. No advantages were found in drip irrigation at high frequency (daily or
several times daily) compared with low application frequency (once or twice a week).
More recent experiments have shown that drip irrigation at the same frequency as
sprinkler (every 2-3 weeks) still has some advantages. The increased yield obtained
with drip was attributed to several factors:
1. Better water distribution, i.e., less variability in water application
2. More control on precise water and fertilizer application
3. More flexibility in irrigation; larger areas can be irrigated with the same water dis-
charge rate and at a lower pressure compared with sprinkling
4. Irrigation can be carried out at all hours, independent of wind
5. Less labor-intensive, once the system is installed
6. More flexibility in insect and weed contro!.
Table 6.5 gives the results of surveys carried out in commercial farmers' fields by
extension officers, showing a consistent yield increase for drip and a gradual replace-
ment of sprinkling by drip. An economic analysis of cotton irrigation technologies
has shown that, under Israeli conditions, drip irrigation is more efficient than sprin-
134 Practical Applications of Drip Irrigation

kling, linear movement or center pivot, even without taking into consideration the
possible yield increase with drip (Rymon and Fishelson 1988).
Drip irrigation of cotton with water of different salinity (3.7 vs 7.8 dS/m) was com-
pared with sprinkler with irrigation amounts of 230-780 mm at different irrigation
intervals and for two intra-row spacings (Meiri et al. 1992). It was found that all data
fell on a single linear response curve relating seed cotton yield to the sum of applied
water and the depletion in soil stored water. With drip irrigation larger amounts of
water could be applied without drainage below the root zone, resulting in higher yields
(Meiri et al. 1992).
Subsurface drip irrigation was compared to surface drip irrigation for cotton in
Israel by Plaut et al. (1985). They found similar yields for both systems when adequate
water was supplied. In the case of deficit irrigation, however, subsurface drip was supe-
rior since most of the soil evaporation was eliminated (Plaut et al. 1985).

6.2.2
Other Row Crops

Drip irrigation is suitable for row crops, but in many cases the investment for its instal-
lation may be relatively high while the economic advantage over other methods may
not be apparent. For example, 10000m of lateral tubing is required for a field of
1 ha with 1 m row spacing in which a lateral is placed next to each row of plants. In
most cases, this high expenditure cannot be compensated by increased crop yields.
Therefore, the prevalent drip system in row crops consists of irrigating every second
row, with some changes in row spacing if necessary.
Experiments carried out with sorghum in lysimeters showed no significant differ-
ences in yield or water use efficiency for drip and sprinkler at high frequency (Ravelo
et al. 1977). They also found that variation in drip irrigation frequency (once, twice
or three tim es a week) had no effect on sorghum yield and water use. Bucks et al.
(1982) present data showing the same lack of crop response to increased frequency of
drip irrigation. Similar results were obtained with corn both in Texas and Kansas.
Daily or weekly, sub- or top-irrigated corn gave similar yields if adequate water was
applied (HoweIl et al. 1997; Lamm et al. 1995). High irrigation frequency may result
in shallow roots, and a higher crop drought sensitivity. In cases of system breakdown
(i.e., mechanical or power failure of the pumping system), the crop irrigated at high
frequency will suffer most. A permanent drip irrigation system makes frequent irri-
gation possible. The frequency should be adapted to the soil (depth and water-holding
capacity), the crop (root system development and drought sensitivity), the climate
(evaporative demand), and the water quality (salinity). Circumstances under which
drip seems to be especially effective include supplemental irrigation in areas with
erratic rainfall. Phene and Beale (1976) showed that drip at high frequency gave higher
corn yield and less water use than furrow or sprinkler irrigation at the same fre-
quencies. With furrow or sprinkler irrigation, no fine control on water application is
possible, and, thus, together with rainfall, excess amounts of water may be applied,
resulting in soil aeration problems. Fertilizers should be applied to the irrigation water
at an adequate rate. A system of twin-row spacing, developed for these circumstances,
requires 40% less irrigation tubing than conventional spacing (Phene and Beale 1979).
For other row crops, as with cotton, water saving with drip irrigation compared with
Field and Fodder Crops 135

other systems is primarily implemented by high water application efficiency and

better control.

6.2.3
SugarCane

Sugar cane in Hawaii is an example of a crop where drip is taking over as the princi-
pal irrigation method. In 1975, less than 10% of the sugar-cane area in Hawaii was
irrigated by drip, while in 1979 more than 40% of the total area was drip irrigated.
The forecast for 1985 was that 80-90% of all the plantations will convert to drip (Bui
and Kinoshota 1985). This rapid conversion has occurred because of cost savings,
yield increase and water conservation (Gillespie 1980). Gibson (1976) summarized the
early development of drip irrigation of sugar-cane. In 1940, the furrow-irrigated cane
fields star ted to be converted to sprinklers in order to reduce operating costs. In 1970,
when above-canopy sprinkling was practiced on about 10% of the cane area, the devel-
opment of drip irrigation began. The cane crop is harvested 24 months after plant-
ing, at which time the drip laterals are discarded. In general, disposable twin-wall
tubing is used, with laterals placed between two dosely-spaced plant rows (91 cm
apart), while the next pair of rows is 183 cm away. The system is operated 12 h daily,
applying 8-10mm water. The dogging problem was solved by adequate filtration and
chlorination (Gillespie 1980). Drip irrigated fields have yielded 15-25% more than
furrow irrigated plots, and the capital costs for drip are lower than those for sprin-
kler installations. Lower water application and labor requirements, high er water and
nutrient efficiencies and improvements in other operations, such as weed contro!, are
additional reasons for the rapid development of drip for sugar-cane in Hawaii (Bui
and Kinoshota 1985). In other sugar-producing areas, the same development is taking
place, although at a slower pace.

6.2.4
Forage Crops

Drip irrigation is not practical for most irrigated forage crops because the complete
cover of the crop requires relatively dose spacing of the laterals, making installation
very costly. Furthermore, the harvesting of the forage may damage the irrigation
equipment. An exception is the growing of corn as a forage crop on small farms, as
practiced in Israel. Corn can give very high yields of dry matter (20-30 Mg/ha) which
makes it a very attractive forage crop. The use of overhead sprinklers for corn in small
plots is cumbersome, since the height of the sprinklers has to be adjusted as crop
height varies. Some of the water is lost at the borders of the plots and wind interferes
with irrigation. The drip system for forage corn developed by Leshem (1981) divided
the field into 0.5-1 ha subplots, which are plan ted consecutively at 20-day intervals.
The first irrigation is carried out by sprinkler. Then, between 20 and 65 days after
planting, the crop is irrigated by drip every 3 to 4 days. The distance between the lat-
erals is 100-200 cm, depending on the soil and planting pattern. Before cutting the
crop after 65 days, the laterals are moved to the next plot. Each set oflaterals can irri-
gate three to four plots per season. The amount of water (to which fertilizer is added)
for each irrigation is determined by pan-evaporation, the pan factor being adjusted
136 Practical Applications of Drip Irrigation

to the growing stage of the crop. The dry matter yield of forage corn obtained with
this drip system was similar to that obtained with sprinkler irrigation but with a lower
water application (180mm) than with sprinkler (450mm) (Leshem 1981).

6.3
Vegetables

Vegetables are high-value crops, generally very sensitive to water stress, especially
with respect to yield quality. Moreover, most vegetables are shallow-rooted (Carr,
1981). Therefore, any irrigation system providing a constant supply ofwater without
the development of water stress is suitable for vegetable growing. Drip irrigation is
most suitable for vegetable growing since it does not interfere with harvesting oper-
ations, fertilizers can be supplied through the system, foliage is not wetted, and water
can be supplied frequently and accurately. With frequent drip irrigation, use ofbrack-
ish water is possible. A relatively low salt concentration is maitained in the root zone,
while avoiding foliar damage (Bernstein and Francois 1973). Many experiments with
vegetable crops have been reported, of which only a few will be discussed to empha-
size a few major points. Lettuce is a very shallow-rooted crop and can give high yields,
provided water can be applied frequently. Comparable lettuce yields were obtained
with drip, sprinkler, sub-surface and furrow irrigation at high frequency (Sammis
1980). With potatoes, the highest yields were obtained with subsurface irrigation, com-
pared with the other three methods, which gave similar yields (Sammis 1980).
Shalhevet et al. (1983) obtained similar potato yields under drip and sprinkler irri-
gation with the latter requiring 10% more water. They attributed the higher water
requirement to evaporation during sprinkling.
One of the first experiments with drip irrigation was carried out in Arizona with
cabbage (Bucks et al. 1974). It was found that the consumptive use of water with drip
and furrow irrigation at 12-day intervals was the same, and yields were similar.
Increasing drip frequency to every 6 or every 3 days gave no increased yield, whereas
application of water at less than consumptive use decreased production. With toma-
toes, on the other hand, although no differences were observed between drip and
furrow irrigation at the same frequencies, yields were higher when water was applied
more frequently (Freeman et al. 1976). The effect of irrigation frequency and subse-
quent development of water stress seems to be specific for each crop and related to
soil properties. For beans it depends on the plant growth period; irrigation frequency
during the pre-flowering period did not influence pod yield. The highest yield was
produced under frequent drip irrigation after flowering. Weekly furrow irrigation
during pod set also resulted in a yield increase over bi-weekly irrigation (Muirhead
and White 1981). For tomatoes growing on a sandy desert soil, daily drip resulted in
much higher yields and growth than did sprinkler irrigation every 3 days. These dif-
ferences could be attributed to marked lowering of soil and plant water potential
(Goldberg et al. 1976). In strawberry culture, the drip laterals can be placed below the
plastic mulch, which is usually placed underneath the plants. This practice results in
considerable water saving and increased production compared with overhead sprin-
kling. Locascio et al. (1977) have shown that for best results at least 50% of the fertil-
izer should be applied with the irrigation water. Some of their results are shown in
Table 6.6.
Vegetables 137

Table 6.6. Effect of irrigation method and timing of fertilizer application on yield of mar-
ketable strawberries (Locascio et al. 1977)

Irrigation method No/ha (x10 3 ) Yield (Mg/ha) Size (g/fruit)

Drip (daily) 1779 18.85 10.64

Overhead sprinkle (weekly) 1755 16.83 9.77
Control (rainfall only) 1379 13.80 10.00
Fertilizer (%) applied with drip
0% 1508 16.49 10.91
50% 1902 20.47 10.76
100% 1794 18.42 10.38
Frequency of fertilization
Daily 1806 19.50 10.78
Weekly 1888 19.39 10.37

6.4
Subsurface Irrigation

In its early development, drip irrigation used laterals with emitters buried in the soi!
(Blass 1971). This practice was soon discarded because of clogging and root penetra-
tion problems. Recently, this practice has been revived for several reasons:
1. The development of more reliable emitters less prone to clogging and new solu-
tions to root penetration problems (trefian)
2. The possibility of using the same system for several consecutive crops, thus saving
costs
3. Reducing the labor requirements involved in removing and re-installing the tubing
for each crop
4. Less interference with weeding, spraying or harvesting of the crops by buried lat-
era
s than by the surface-placed laterals
5. Minimizing water loss by evaporation from the soi! surface.
Successful experiments with subsurface drip irrigation were carried out by Bucks et
al. (1981). They found no appreciable differences in yields of melon between surface
drip, sub-surface drip and furrow irrigation, provided water was applied according to
seasonal evapotranspiration demand. Successful crops of onions and carrots were
produced with the same subsurface drip system after two previous consecutive crops
of melons. No problems of root penetration into the subsurface emitters were encoun-
tered. Phene (1995) carried out an experiment comparing high-frequency subsurface
drip irrigation (several irrigations per day) with high-frequency surface drip and with
low-frequency surface drip on the yield of processing tomatoes. Some of his results
are shown in Table 6.7. These data show that injection of a complete fertilizer is essen-
tial with high-frequency drip irrigation; yields were almost doubled in 1987, com-
pared with 1984, when only nitrogen was applied. Evapotranspiration (ETJ was
measured continuously by a weighing lysimeter in the sm treatment and irrigation
was applied accordingly. Bare soi! evaporation was 0.06ET, for the sm treatment com-
pared with 0.12ET, for the HFDI treatment. The SDI system has been in operation for
more than 10 years. The lack of root intrusion and the maintaining of the original dis-
138 Practical Applications of Drip Irrigation

Table 6.7. Fresh yield of tomatoes and water use efficiency as effected by irrigation method
(Phene 1995)

Treatments 1984 (N only) 1985 (N + P) 1987 (N + P + K)

Yield (t/ha) WUE Yield (t/ha) WUE Yield (t/ha) WUE

SDI 121a 18a 168a 22a 220a 31a

HFDI 126a 19a 152b 20b 201b 29b
LFDI 114a 16b 130c 18c 187c 26c

WUE = fresh yield/crop evapotranspiration (kg/m 3 ), SDI = high-frequency subsurface drip

irrigation, HFDI = high-frequency surface drip irrigation, LFDI = low-frequency surface drip
irrigation

charge of the system was attributed to the high frequency of irrigation, continuous
injection of phosphorous acid and yearly injection of fumigant (Phene 1995).
The frequency of SDI had no effect on com yield in Texas and in Kansas; similar
yields were obtained with daily and weekly irrigations (Caldwell et al. 1994; Howell et
al. 1997; Lamm et al. 1995). A more efficient use of nitrogen can be achieved with SDI
rather than by surface drip irrigation (Lamm et al. 1995). Indications of high er P and
K uptake by sweetcom, using subsurface drip fertigation, as compared with surface
drip fertigation in a container experiment were reported by Martiez Hernandez et al.
(1991). Enhanced root development in the region of high nutrient concentrations was
assumed to be the reason for this. Enhanced Puptake with subsurface drip fertiga-
tion was not obtained in a field experiment (Bar-Yosef et al. 1989).
Another phenomenon sometimes observed with SDI is a decrease in the dis charge
of the emitters because of apressure build-up in the soil (Shani et al., 1996). In soils
with low hydraulic conductivity and when the soil is compacted around the emitter
this phenomenon may become appreciable, especially with emitters with a high
nominal discharge. In light-textured soils and when there is a cavity around the buried
emitter, the decrease in emitter dis charge is small (Shani et al 1996).