Water is becoming increasingly scarce worldwide. Aridity
and drought are the natural causes of scarcity. More recently however,
man-made desertification and water shortages have aggravated natural scarcity
while at the same time population is increasing and there is increased
competition for water among water user sectors and regions. Rainfall is
not sufficient in many regions and predictions on climate change show
that problems are likely to increase; thus available water resources are
increasingly limited in quantity. In addition, the quality of water is
often degraded, so that water resources become less and less available.
Irrigated agriculture is therefore forced to find new approaches to meet
the demands of water scarcity, environmental friendliness, economic viability
and social equilibrium (Pereira, 2006).
Long term average annual precipitation in the Aegean region
is about 657 mm, with more than 89% of it falling from October to March.
Water loss by evapotranspiration is very high during the growing season.
Therefore, irrigation is needed during the growing season to maintain
and enhance crop growth and yield. Also, irrigation water is the most
important limiting factor for agriculture during the hot and dry summer
period of the region. Limited availability of irrigation water requires
fundamental changes in irrigation management or the application of water
saving methods. A generally applicable procedure is to assess the benefits
of changing irrigation water management based on deficit irrigation, which
is the practice of deliberately under-irrigating field crops. Under these
conditions, there is one way for farmers to maximize their profit from
maize production. This is to determine the water-yield relationships of
maize crops and to choose the most appropriate irrigation scheduling in
order to conserve irrigation water. In this way, optimum irrigation schedules
for maize should be determined in order to cope with prevailing conditions
and unplanned water shortages in the region. Knowledge of the sensitivity
of maize to water stress over the whole growing season or at one of the
different growth stages has been widely used in studies aiming to develop
deficit irrigation strategies, as well as to determine the yield response
factor (ky) of maize.
This is a parameter used to quantify the effect of water
stress, derived from the linear relationship between relative seasonal
evapotranspiration deficits (1-ETa/ETm) and relative
yield loss (1-Ya/Ym) (Dagdelen et al., 2006;
Yazar et al., 2002; Musick and Dusek, 1980; Doorenbos and Kassam,
Maize is a major commercial field crop in the Aegean region
of Turkey. It has become a widely grown feed grain crop particularly as
a second crop after wheat or barley. In Turkey, maize production is about
3,000,000 Mg of grain maize from 545,000 ha (Anonymous, 2006). Maize is
sensitive to water deficit. This sensitivity to water stress means that
when water is limited it is difficult to implement irrigation management
strategies without incurring important yield losses (Lamm et al.,
1994; Farre and Faci, 2006).
Several experimenters have subjected maize to a water deficit
during different developmental stages. It was found that both the time
and the degree of stress are important in determining the final grain
yield. It is well established that a water deficit in the period which
includes anthesis can have a disastrous effects on the maize grain yield,
whereas the effects of moisture stress are less drastic at other growth
stages (Moser et al., 2006). Irrigating a crop with the required
quantity of water during the moisture sensitive period of flowering and
yield formation stages, yet allowing moderate stress at vegetative and
maturity stages produce the optimum yield with maximum water use efficiency
and water economy in most crops (Panda et al., 2004; Shaozhong
et al., 2000).
Many studies have shown that the relationship between maize
yield and seasonal crop water use is linear (Dagdelen et al., 2006;
Payero et al., 2006; Cetin, 1996; Howell et al., 1995; Cosulluela
and Faci, 1992). On the other hand, predicting the yield response of maize
to water use is important in developing strategies and decision-making
for farmers and their advisors and research for irrigation management
under limited water conditions.
The objective of this study was to determine water-yield
relations of maize and the effects of limited water on yield, yield response
factor, water use efficiency, irrigation water use efficiency, dry matter
and leaf area index. The results of this study will provide a guideline
to regional growers and irrigation agencies on water-saving irrigation
and optimum water management programs for maize in the Aegean region of
MATERIALS AND METHODS
Site and climate: Field data for this study were collected in 1999 and 2000
at the Agricultural Research Station of Aegean University, Izmir, Turkey
(latitude 38° 28` N, longitude 27° 15` E, altitude 27 m).
The soil of the experimental site is Sandy Clay Loam (SCL) with water
content at field capacity varying from 18.81 to 22.94% and wilting point
varying from 8.45 to 10.72% on a dry weight basis. The physical and chemical
properties of the soil in the experimental crop field are given in Table
1 and 2.
||Physical properties of the soils of experimental
The climatic variables for the experimental years were recorded
at a weather station located close to the experimental site. These variables
and the long-term trends for the growing season (June-November) are shown
in Table 3. The experimental area has favourable soil
and climate conditions for maize production.
Irrigation water applied during the experimental years was
also analysed (EC:1.2 dS m-1; pH: 6.79; SAR:1.01) and classified
C3S1. According the EC, pH and SAR values, it can
be concluded that the water used in irrigation is proper for maize production.
Crop agronomy: In 1999, maize was planted on July 20 and harvested on November
11 (Day of Year, DOY:201). In 2000, it was planted on July 10 and harvested
on November 1 (Day of Year, DOY:193). Maize plants were thinned to a spacing
of 0.70 m (row width)*0.20 m. Weeds, pests and diseases were controlled.
Maize plots were fertilized with 50 kg day-1 pure NPK (20:20:0)
before sowing and first irrigation. Plant density is 7.1 plant m-2.
Irrigation treatments: The experiment was conducted using a randomized complete
block design with three replications. Each experimental plot was designed
as 10 m long by 5.0 m wide. There was a 2.0 m space between each plot
in order to minimize water movement between treatments.
In this study, five irrigation treatments, differing in
irrigation rate, were evaluated at ten day intervals. Irrigation treatments
were tested with 100, 70, 50, 30 and 0% replenishment of water depleted
at 120 cm soil profile from 100% replenishment treatment. Each year, treatments
included a dryland treatment (I0) which received no irrigation.
This many treatments were included to obtain enough data points and a
wide enough range of water stress levels to be able to develop meaningful
quantitative relationships between irrigation, yield and other parameters.
Also, irrigation was applied when approximately 50% of the available soil
moisture was consumed in the effective root zone at the control treatments,
called I100. The measured soil moisture level in the I100
treatment was used to initiate irrigation of maize during the growing
season. In treatments, I70, I50, I30 and
I0 irrigations were applied at the rates of 70, 50, 30 and
||Chemical properties of the soils of experimental
||Long-term monthly climatic data for the growing
season at the experimental site
of control treatments (I100) on the same day,
respectively. The closed-end furrow irrigation method was used in all
treatments. Water applied to each experiment plot was measured with water
metered connected to an irrigation pipe. Irrigation was started on 13
August 1999 and 11 August 2000.
Soil and crop measurements: Soil water content was monitored gravimetrically in each
0.3 m layer down to a depth of 1.20 m for each treatment before each irrigation
Crop evapotranspiration (ET) was calculated using the soil
water balance equation for the growing season as follows (Heermann, 1985):
||Irrigation application (mm)
||Drainage (mm) and
||The change of soil water storage at the measured soil depth.
Runoff and drainage were considered negligible in the water balance.
The effect on yield of water stress during the growing season
was calculated as follows (Doorenbos and Kassam, 1979):
||Actual harvested yield (kg ha-1)
||Maximum harvested yield (kg ha-1)
||Yield response factor
||Actual evapotranspiration (mm)
||Maximum evapotranspiration (mm)
||Relative yield decrease
||Relative evapotranspiration deficit
Water Use Efficiency (WUE) and Irrigation Water Use Efficiency
(IWUE) were calculated as follows (Howell et al., 1995):
||Grain yield (kg ha-1)
||Grain yield for equivalent dry land
||Amount of irrigation applied for level i. In most cases, Io
would be zero
Crop development stages in maize were recorded in all treatments.
A phenological stage was defined as 50% of the plants reaching that stage.
Regular observations were made of leaf area index and dry matter. Collections
of maize plant samples were started after first irrigation and continued
until harvest. Maize leaves were separated from the stem and the leaf
area of plants was measured using a scanner with FLAECHE packing programme.
Maize leaves and the rest of the plants were cut into pieces and then
oven dried at 65°C to a constant weight (Gardner et al., 1985;
Analysis of variance (ANOVA) was used to evaluate the effects
of the treatments on the yield, LAI and DM components. Mean comparisons
were made by the LSD (least significant difference) method with p< 0.05.
The analyses were conducted using the TARIST program (Acikgoz et al.,
RESULTS AND DISCUSSION
Water-yield relationship: A total of six irrigations were applied to maize in all
treatments during the growing season. As shown in Table
4, the amount of irrigation water applied varied from 96.8 to 323.2
mm in 1999 and from 139.9 to 466.6 mm in 2000. The seasonal values of
crop water use per treatment ranged from 142.1 to 481.9 mm in 1999 and
from 136.2 to 599.4 mm in 2000. The highest seasonal crop water use occurred
in the full irrigation treatment (I100) owing to an adequate
soil water supply during the growing season and the lowest crop water
use occurred in the non-irrigated treatment (I0). Crop water
use values were affected by irrigation treatments and years. These differences
can be attributed to climatic factors and irrigation scheduling practices.
Seasonal crop water use of maize obtained by Kanber et al. (1990)
was 474.2-605.8 mm in the Cukurova region, while Istanbulluoglu and Kocaman
(1996) obtained 353-586 mm in the Thrace region. In addition, Tolk et
al. (1998) obtained 357-587 mm, Katerji et al. (1996) 494-644
mm, Dagdelen et al. (2006) 169-547 mm and Igbadun et al.
(2006) 385.4-537.1 mm. Also, crop water use for maize without water deficit
was reported by Pandey et al. (2000) as 641-668 mm, while Stegman
(1986) reported 432-514 mm. The results observed in this research were
in agreement with that given above.
Irrigation treatments also resulted in differences in grain
yield as shown in Table 4. This ranged from 3750 to
10639 kg ha-1 in 1999 and from 2136 to 10383 kg ha-1
in 2000 for the different irrigation regimes. Increased water
Amount of irrigation water,
crop water use and grain yield
for the years 1999 and 2000
Relationship between grain
yield (Y) and seasonal crop water use (ET) for maize
amounts resulted in a relatively higher yield, since water
deficit was the main yield-limiting factor in both years. The maximum
yield was obtained at I100 and the minimum yield at I0
in both 1999 and 2000. Grain yields from the experiments were considered
adequate as they compared well with the world average grain yield of maize
of 2004 of 4907 kg ha-1 (Fao, 2006).
Under the conditions of the Harran plain, Cetin (1996) reported
the highest yield of 10150 kg ha-1 using the furrow irrigation
method. Dagdelen et al. (2006) in Western Turkey found that the
average maize yield varied from 2880-11340 kg ha-1 and that
the highest average maize yield was obtained from full irrigation treatments.
Cakir (2004) determined grain yield of maize as 3147-12438 kg ha-1
and Tolk et al. (1998) found it to be 4110-8480 kg ha-1.
When the water saving in this study was 30% (I70), 50% (I50),
70% (I30) and 0% (I0) of I100, the rates
of decrease in maize grain yield were found to be 10.5, 39.2, 45.2 and
64.8 of I100 in 1999 and 27.4, 47.1, 59.5 and 79.4 of the I100
in 2000, respectively. The results indicated that deficit irrigation
affected grain yield significantly.
The relation between seasonal crop water use and grain yield
have been evaluated for each year (Fig. 1)
The relationship between seasonal crop water use and grain yield was
linear for each experimental year. Their
Relationship between grain
yield (Y) and irrigation water (IW) for maize
||Water use efficiency values for the experimental
relationship was significant at p<0.05. A linear relationship
between crop water use and yield for maize has been reported by other
researchers (Payero et al., 2006; Dagdelen et al., 2006;
Cetin, 1996; Howell et al., 1995; Cosulluela and Faci, 1992).
Also, the relation between grain yield and seasonal irrigation
applied to maize was evaluated for each experimental year (Fig.
2). The relationship between grain yield and irrigation applied was
also linear (p<0.05) in 1999 and 2000.
Farre and Faci (2006) also observed grain yield to be linearly
related to seasonal irrigation applied to maize. The findings of their
work showed that in the northeast of Spain and on a loam soil, the yield
penalties of moderate or severe water deficit are more important in maize
than in sorghum.
Water use efficiency and irrigation water use efficiency: Water
Use Efficiency (WUE), expressed as the ratio of grain yield to seasonal
crop water use, was affected by the irrigation treatments in maize (Table
5). WUE values ranged from 1.49 to 2.71 kg m3 both years.
In 1999, WUE for I30 treatments was the highest, while that
for I50 treatment was the lowest. On the other hand, no significant
difference was found between treatments I100 and I70
in either year. In 2000, I100 was the highest and I30
was the lowest. However, a wide range of WUE values have been found
for maize in different studies (Farre and Faci, 2006; Yazar et al.,
2002; Tolk et al., 1998; Steele et al., 1997; Koksal, 1995;
Steele et al., 1994; Caldwell et al., 1994).
The relationship between relative
yield decrease and relative evapotranspiration deficit for growing
Irrigation Water Use Efficiency (IWUE) values, expressed
as the ratio of grain yield to total irrigation water applied, varied
from 1.44-2.55 kg m-3. In 1999 the IWUE of treatment I70
was the highest and the lowest IWUE occurred with treatment I50.
Similarly in 2000, the highest IWUE was obtained from I100
treatment and the lowest IWUE was obtained from I50 treatment.
These results were similar to other values reported for maize. Musick
and Dusek (1980) reported IWUE for maize between 2.44-2.70 kg m-3;
Howell et al. (1995) found that IWUE was between 1.51-2.48 kg m-3;
Caldwell et al. (1994) determined these values as 2.07-2.76 kg
Relative yield decrease-relative evapotranspiration deficit
relationship and yield response factor (ky): The relationship between relative yield decrease and relative
evapotranspiration deficit is shown in Table 6 and yield
response factor (ky) is shown in Fig. 3. The slope of
the fitted regressions represents ky.
The ky values of maize to water deficit for the entire growing
season were 0.90 in 1999 and 1.07 in 2000. When the values of the two
experimental years were combined, the coefficient of determination (r2)
was 0.939. According to the regression equations, ky was 0.99. Generally,
the ky value obtained in this study was consistent with those reported
by Koksal and Kanber (1998) as 1.03, by Gencoglan (1996) as 1.23, by Cakir
(2004) as 1.29, by Dagdelen et al. (2006) as 1.04 and by Karam
et al. (2003) as 0.81. Some differences could be explained by the
high relative humidity and different precipitation characteristic of the
coastal areas. On the other hand, Igbadun et al. (2006) reported
the ky value to be 1.90. The high value for ky obtained in their study
is an indication of severe moisture stresses or low resistance to moisture
stress. It implies that the rate of relative yield decrease resulting
from moisture stress is proportionally higher than the relative evapotranspiration
Above-ground dry matter (DM): Above-ground Dry Matter (DM) was also significantly affected
(p<0.05) by water deficit in both 1999 and 2000 (Fig. 4,
The highest level of dry matter of maize, obtained from
I100 treatment, was 3.375 and 3.015 kg m-2 and
Yield response factor, ky,
for maize in 1999 (a), 2000 (b) the two experiments combined
||Dry matter (DM) for maize in 1999
the lowest, obtained from I0 treatment, was 0.709 and 0.573
kg m-2 in 1999 and 2000 respectively. A significant difference
was found between all treatments (p<0.05).
||Dry matter (DM) for maize in 2000
The relationship between dry
matter yield (Y) and irrigation water (IW)
The relationship between dry
matter yield (Y) and crop water use (ET)
Increasing the amount of irrigation appeared to increase
DM. That is, water deficit reduced dry matter. The results were consistent
with those of Farre and Faci (2006). In addition, DM was found to be 2.5
kg m-2 by Karam et al. (2003) and 3.25-3.45 kg m-2
by Dagdelen et al. (2006). There was also a significant linear
relationship between DM and irrigation applied and DM and crop water use
obtained for maize, as shown in Fig. 6 and 7.
Leaf area index (LAI): Results obtained from the two experimental
years showed a significant effect of irrigation application on LAI (Fig.
8). The highest LAI, obtained
from I70 treatment, was 4.84 in 1999 and from
I100 treatments, 5.31 in 2000. The lowest LAI was recorded
from non-irrigated conditions, treatment I0, in both years.
These results showed that water deficit causes a decrease in LAI and reduction
of yield. This result was consistent with the finding of Istanbulluoglu
and Kocaman (1996) and Cosulluea and Faci (1992). Howell et al.
(1995) determined that the highest LAI was 4-5.5 for maize under non-water
stressed conditions. Maximum LAI of maize was reported to be 5.03-5.35
under well irrigated conditions and as 3.43-3.0 under non irrigated conditions,
by Gencoglan (1996). Cakir (2004) reported that LAI increased under favourable
soil moisture conditions until 70-80 days after emerge and then decreased
as the older leaves died. In general the results obtained in this study
are similar to those reported by other researchers. However, the slight
differences could be due to differences in techniques used in the agronomic
process, differences in climate between locations especially the amounts
and distribution of precipitation, differences in crop varieties, cultural
practices, irrigation methods and irrigation scheduling practices.
It was concluded from this study that maize grain yields
were significantly affected by irrigation applied during the course of
the growing season in 1999 and 2000. In both experimental years, grain
yield ranged from 2136 to 10639 kg ha-1 with respect to the
irrigation treatments and other components decreased according to the
water deficit levels created.
Yield response factor (ky) was estimated as 0.99
when the experimental years were considered together. ky obtained
for this study could be used for the purposes of irrigation management
and water allocation scheduling over irrigation schemes under limited
irrigation water supply. Water use efficiency values varied from 1.49
to 2.71 kg m-3 and irrigation water use efficiency values varied
from 1.44 to 2.55 kg m-3. Dry matter yield (DM) and Leaf Area
Index (LAI) increased with increased water use in both experimental years.
Maximum measured LAI was 4.84 for the treatment I70 in 1999
and 5.31 for the treatment I100 in 2000. Minimum LAI was obtained
for treatment I0 in 1999 and 2000. As the crop water stress
increased, LAI values decreased due to a reduced size of the leaves. A
positive linear relationships between crop water use and yield exists
during the experimental years.
The finding of this study showed that treatment I100,
designated to receive 100% soil water depletion every ten days, could
be used for maize grown in semi arid regions under no water scarcity.
On the other hand, the treatment of 30% water saving (I70)
reduced maize yield by 18.96%. Thus, results obtained from treatment I70
could be used as a good basis for deficit irrigation strategy development
in regions where irrigation water supplies are limited. However, severe
deficit in irrigation water amounts will cause significant declines in
The research was supported by The Scientific Technological
Research Council of Turkey (TUBITAK-TARP-2340) and Ege University Scientific
Research Foundation (99-ZRF-031), Turkey. The author is grateful to these
organizations for their valuable assistance.