Agronomy Journal Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (35)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Agricola
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Related Collections
Right arrow Agroclimatology
Right arrow Water Use
Right arrow Wheat
Agronomy Journal 92:231-238 (2000)
© 2000 American Society of Agronomy

CLIMATOLOGY

Water Use Efficiency of Rainfed and Irrigated Bread Wheat in a Mediterranean Environment

Theib Oweisa, Heping Zhanga and Mustafa Palaa

a International Center for Agricultural Research in the Dry areas (ICARDA), P.O. Box 5466, Aleppo, Syria

t.oweis{at}cgiar.org


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results
 Discussion
 REFERENCES
 
In West Asia and North Africa, shortage of water limits wheat (Triticum aestivum L.) production. Current irrigation practices aim at maximizing grain yield, but achieve lower return for the water consumed. Maximizing water use efficiency (WUE) may be more suitable in areas where water, not land, is the most limiting factor. We examined the effects of various levels of supplemental irrigation (SI) (rainfed, 1/3 SI, 2/3 SI, full SI), N (0, 5, 10, 15 g N m-2), and sowing time (Nov., Dec., Jan.) on evapotranspiration (ET) and WUE of wheat. WUE was calculated for rain (WUEr), for total water (gross: rain + irrigation) (WUEg), and for SI water only (WUESI). ET ranged from 246 to 328 mm for rainfed crops, with grain yield ranging from 130 to 270 g m-2 and total dry matter from 380 to 1370 g m-2. Irrigated crops had ET of 304 to 485 mm, with grain yield of 170 to 500 g m-2. The degree to which water supply limits grain yield was indicated by the ratio of pre- to post-anthesis ET (2.1–2.4:1). The SI treatments significantly increased WUEg: from 0.77 to 0.83 to 0.92 kg m-3 in November and December sowings for 1/3 SI and from 0.77 to 0.92 kg m-3 in November sowing for 2/3 SI. The highest WUEg and WUESI were achieved at 1/3 to 2/3 SI. WUE was substantially improved by applying 5 and 10 g N m-2, with little increase for higher rates. Delaying sowing had a negative effect on WUE for both irrigation and rainfed conditions. In this rainfed Mediterranean environment, WUE can be substantially improved by adopting deficit SI to satisfy up to 2/3 of irrigation requirements, along with early sowing and appropriate levels of N.

Abbreviations: ET, evapotranspiration • SI, supplemental irrigation • GY, grain yield • TDM, total dry matter • WANA, West Asia and North Africa • WUE, water use efficiency • Subscripts: g, gross • r, rainfed • SI, supplemental irrigation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results
 Discussion
 REFERENCES
 
IN WEST ASIA AND NORTH AFRICA (WANA), water resources are generally scarce, and agriculture's share of these resources is declining due to competition from domestic and industrial sectors. In this region, a typical Mediterranean climate prevails, with rain falling mainly during the winter (and a lesser amount during the warmer spring period); this rainy season is followed by a hot, dry summer. Rainfed crop production under this climate thus depends strongly on both the amount and distribution of rain. In the WANA region, amount of rainfall is low and generally poorly distributed, so periods of water deficit occur during the grain-filling stage of wheat almost every year (Oweis et al., 1992). As a result, crop yield and water use efficiency (WUE) are generally low and variable. The production of 1 kg of wheat (T. aestivum L.) grain under fully irrigated conditions requires about 1 to 2 m3 of irrigation water (Perrier and Salkini, 1991); in rainfed areas it requires from 1 to 3 m3 of rainwater (Cooper et al., 1987a; Perrier and Salkini, 1991). Since water is the major limiting factor for agriculture in the WANA region, improving WUE is vital for meeting the increasing food demand (Cooper et al., 1987b).

Supplemental irrigation (SI) is defined as the application of a limited amount of water to rainfed crops when precipitation fails to provide the essential moisture for normal plant growth. This practice has shown potential in alleviating the adverse effects of unfavorable rain patterns and thus improving and stabilizing crop yields (Perrier and Salkini, 1991; Oweis et al., 1998; Zhang and Oweis, 1999). Early studies at ICARDA showed that applying two or three irrigations (80–200 mm) to wheat increased crop grain yield by 36 to 450%, and produced similar or even higher grain yields than in fully irrigated conditions (Perrier and Salkini, 1991; Oweis, 1994). Supplemental irrigation is widely practiced in Syria, and in southern and eastern Mediterranean countries. However, excessive use of water in SI because of low irrigation cost and attractive gains from increased yields has resulted in a decline of aquifers and deterioration of water quality in many areas (Ward and Smith, 1994).

Increasing the portion of water used for plant transpiration through a large and early canopy can increase WUE. In Mediterranean environments, where crop canopy development in winter is slow and rain occurs as frequent and small events, soil water evaporation may account for 30 to 60% of seasonal ET (Cooper et al., 1983; French and Schultz, 1984; Siddique et al., 1990; Zhang et al., 1998). Thus, agronomic practices that reduce soil water evaporation via a larger plant canopy and early ground cover and at the same time increase the crop's ability to extract soil water may increase the amount of water transpired and, consequently, WUE. Nitrogen deficiency is another major constraint in canopy development in the Mediterranean region (Anderson, 1985). Crop responses to N fertilization depend on the level of water availability (Pala et al., 1996). Application of fertilizers not only increases plant shoot and root growth (Brown et al., 1987), but also increases ET through a larger root system and greater extraction of stored water (Cooper et al., 1987a). In addition, a large and earlier canopy cover resulting from the application of N can reduce soil water evaporation and increase crop WUE (Zhang et al., 1998).

Under rainfed conditions, the date of the first significant rain determines the sowing date. Early sowing of appropriate cultivars is a recognized means of increasing wheat yields in other Mediterranean-type environments, such as Western Australia (Anderson and Smith, 1990; Anderson, 1992). Using a simulation model, Stapper and Harris (1989) estimated that wheat grain yield in Syria declined by 4.2% per week when sowing was delayed after 1 November.

In the water-scarce areas of West Asia and North Africa, water (not land) is the most limiting factor to wheat production. Satisfying crop water requirements, although it maximizes production from the land unit, does not necessarily maximize the return per unit volume of water. Improving water productivity can contribute to water savings, which can be used to irrigate additional lands with higher total production and/or improve the sustainability of the existing water resources. It is assumed that maximum WUE may be achieved at irrigation levels below those that satisfy full crop irrigation requirements. However, the SI level, N rate, and sowing date at which WUE can be maximized under the rainfed conditions of the WANA Mediterranean need to be evaluated before improved management strategies can be devised. Our objective for this work was a better understanding of the effects of applying different levels of these inputs and climate interaction on ET and WUE of bread wheat in northern Syria.


   Materials and methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results
 Discussion
 REFERENCES
 
The study was conducted in ICARDA's research station at Tel Hadya, Syria (36°01' N, 36°56' E) for four growing seasons (1992–1996). The soil is classified as a Calcixerollic Xerochrept. The soil water content of the top 1.0 m soil, measured by a pressure plate, is 48% by volume at -0.033 MPa, and 24% at -1.5 MPa.

The treatments comprised three sowing dates (early November, mid-December, and late January), four N rates (0, 5, 10, and 15 g m-2), and four SI rates (full SI, 2/3 SI, 1/3 SI, and rainfed). Four bread wheat (T. aestivum L.) cultivars were used: Cham 4, Cham 6, Gomam, and Mexipak 65. The plots were replicated three times in a split-split-split-plot design. Dates of sowing were the main plots, cultivars the subplots, SI rates the sub-subplots, and N rates the sub-sub-subplots. The area of individual sub-sub-subplots was 14.2 m2 (2.25 by 6.3 m). The crops were planted in 17.5 cm rows at a seed rate of approximately 300 seeds m-2. Nitrogen application (urea) was split; half at planting and half top-dressed at the early tillering stage. Phosphorus was applied as basal dressing each season as triple superphosphate (46% P2O5) at the rate of 4 to 5 g P m-2. The site alternated each year between two adjacent blocks. The preceding crops were chickpea (Cicer arietinum L.) in the 1992–1993 and 1993–1994 seasons and safflower (Carthamus tinctorius L.) in the 1994–1995 and 1995–1996 seasons.

Water was applied to the treatments using a drip irrigation system; the amount applied was measured using a flow meter. This system was designed to ensure full and uniform water coverage and distribution. Emitters were spaced along polyethylene lines in four groups to automatically provide full SI requirements (full SI), 1/3 and 2/3 of full SI to the various plots. Spacing between laterals was 70 cm and between emitters 17.5, 35.0, and 52.5 cm, corresponding to full SI, 2/3 SI and 1/3 SI treatments, respectively. No emitters were installed in the rainfed treatment. Irrigation was applied to all treatments at the same time, when the root zone of the full SI treatment had lost 50% of its available moisture (defined as the difference in water storage in the root zone between field capacity and wilting point). The amount of water then given to the full SI treatments was calculated to refill the root zone close to the field capacity. The root zone depth over the season was estimated based on the soil water depletion pattern determined by a neutron probe. Details on irrigation and date applied are given in Table 1 . In the 1992–1993 season, 45 mm of water was applied to all treatments for sowing in November because the soil moisture was not adequate for seed germination. A single aluminum access tube was installed to a depth of 1.8 m or, where soil depth was less than this, to the maximum depth possible, in each sub-sub-subplot of one replicate for the cultivar Gomam. The soil water content was monitored at approximately 7- to 14-d intervals using a neutron probe (IH-II, Didcot Instruments, Abingdon, UK). The measurements were made for each 15-cm layer in the soil profile from 15 cm to a depth of 180 cm or, where soil depth was less than this, to the maximum depth possible. The water content in the top 15 cm layer was measured gravimetrically. Evapotranspiration was determined for each sub-sub-subplot from sowing to harvest using the following soil-water balance equation:

(1)
where {Delta}S is the change of soil water storage (mm) measured by neutron probe, P is the precipitation (mm), I is the amount of irrigation (mm), and Dr is the drainage below the bottom of 1.8 m access tube (mm). No surface runoff occurred at any time during the four seasons. Based on previous research at the site (Harris, 1994) showing that the soil water deficit in the deep clay soil before the rain season is much higher than the annual rainfall, even in wet years, we assumed no drainage from rainfall; we verified this finding later in the present study. The same assumption was applied to treatments with deficit irrigation. In these treatments, the amount of irrigation water was determined to partially fill the deficit in soil water measured immediately before irrigation. Knowing that the soil water was monitored to a depth of 1.8 m, well below the wheat root zone, it is believed that this assumption was realistic. However, it was found during the analysis of the data that some drainage must have occurred immediately after irrigation in the full irrigation treatments. To estimate the amount of drainage, we assumed that ET within 2 wk after irrigation was equal to the sum of the daily reference ET calculated using the revised Penman–Monteith equation (Allen et al., 1994). This is because it is unlikely that the crop was water-stressed within this short period after irrigation. Values above reference ET were considered drainage.


View this table:
[in this window]
[in a new window]
  		Table 1 Amount of water applied to the full supplemental irrigation treatments (full SI) and application dates for wheat sown at different dates

 
Total dry matter was measured in a 0.26-m2 sample (3 rows, 35 by 25 cm) and was dried at 70°C for at least 24 h. Grain was harvested from a 7.79-m2 sample in each plot using a combine harvest. Wind speed, incoming solar irradiance, wet- and dry-bulb air temperatures, rain, and Class A pan evaporation were measured in the weather station located at the Tel Hadya research station, a few hundred meters from the experiment site. Monthly air temperature, precipitation from a gauge close to the site, and vapor pressure deficit for the four seasons are presented in Table 2 .


View this table:
[in this window]
[in a new window]
  		Table 2 Mean maximum and minimum air temperature, rain and mean vapor pressure deficit at Tel Hadya in northern Syria (1992–1996).{dagger}

 
Water use efficiency (kg m-3) was determined as the ratio of crop production (grain yield and TDM, g m-2) to ET (mm). Evapotranspiration from the rainfed plots was from rain and previous soil water storage, and the corresponding WUE was defined as WUEr. Evapotranspiration in the SI plots was from both rain and SI water and defined as gross WUE (WUEg). Water use efficiency values (WUEr for rainfed plots, WUEg and WUESI for SI plots) were calculated only for the plots in which soil water content was measured (Goman cultivar only).

Water use efficiency for SI water (WUESI) reflects the productivity per unit SI water used by the crop above that of rain. The increased production in the SI plots relative to that of the rainfed plots was assumed to be due only to water applied in SI. Accordingly, WUE for SI water was calculated as the ratio of the difference in crop yield (grain or dry matter) between SI and rainfed plots to the difference in ET for the same treatments as follows (Bos, 1980):

(2)

In Eq. [2], Y is grain or dry matter yield (g m-2), ET is evapotranspiration (mm), and the subscripts r and t refer to rainfed and treatment (rainfed + SI) growing conditions, respectively.

Analysis of variance was used to test the effect of season, SI, N, sowing date, and their interactions on grain yield, TDM, ET, and WUE. Because ET was measured only in one replicate, the data for the four seasons were combined. Analysis of variance of this experiment was used to assess the statistical significance of main effects of the treatment factors, seasonal variation, interactions among the factors, and associated standard errors of means. A close examination of various higher-order interactions (>=3) showed that either they were statistically insignificant or their contribution to total variability was relatively small. The main factors and interactions significantly affecting grain yield and WUEg were used to develop response functions for grain yield and WUEg using a quadratic relation as follows:

(3)


(4)

where Y is grain yield in g m-2, a0 to a9 and b0 to b9 are the respective regression coefficients, Sd is the delay in sowing (first date of sowing was taken as 0), D is the vapor pressure deficit (kPa) during the grain-filling stage (estimated using daily average dry- and wet-bulb temperatures), N is the amount of nitrogen (g m-2), and W is the sum of rain and irrigation amount (mm). Stored soil water from a previous year was not included in the models because it usually does not contribute to ET in this environment except under fallow conditions. Vapor pressure deficit (D) during the grain-filling period was included in the response model to reflect air temperature and humidity effects on crop grain yield and WUE during the grain-filling stage.


   Results
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results
Discussion
 REFERENCES
 
Climate
Seasonal rain was 283 mm in 1992–1993, 358 mm in 1993–1994, 320 mm in 1994–1995, and 395 mm in 1995–1996 (Table 2). The rain was close to the long-term average (330 mm) in the 1993–1994 and 1994–1995 seasons. The 1992–1993 season was slightly drier, with a long dry spell lasting from mid-March to early May, when late rain brought relief just in time for grain filling. The 1993–1994 season was marked by a heat wave in April and May that adversely affected crop grain yields, but favorable precipitation during the winter saved grain yields from falling below average. The growing season of 1994–1995 started with an exceptionally rainy period from mid-November until the beginning of December followed by normal rain in December and January. There was a rather long dry spell extending from the third week of January until mid-April in this season. The last rains that fell at the end of May and the beginning of June were probably too late to benefit the crops. In the 1995–1996 season, early rain was somewhat below average and irregular, but this was compensated by above-average rain later in the season. The air temperature during the four seasons was normal, except for a heat wave during the 1993–1994 season.

Dry Matter and Grain Yield
Average grain yield and TDM for the various treatments over the four seasons (1992–1993 to 1995–1996) are presented for cultivar Gomam in Table 3 . The analysis and discussion of the influence of factors on grain yield and TDM were presented for the other cultivars in a separate paper (Oweis et al., 1998). Generally, sowing date, SI level, and N rate had a significant influence on grain yield and TDM. Crop grain yield (Y) was linearly correlated with TDM for all the data, with the resulting regression equation:

with .


View this table:
[in this window]
[in a new window]
  		Table 3 Mean wheat grain yield and total dry matter for individual treatments during four seasons, 1992 to 1996, by N level applied (0–15 g N m-2)

 
Evapotranspiration
Evapotranspiration was significantly influenced by SI, season, N, and sowing date (Tables 4 and 5) . Sowing date had a significant effect on ET in the 1992–1993 and 1994–1995 seasons (data not shown) but not in 1993–1994 and 1995–1996. Rainfed crops had the lowest ET, which ranged between 246 to 328 mm. This was slightly less than the total rain during the crop growing seasons. Under SI, ET increased as the applied water increased. The range of ET was 338 to 382 mm at 1/3 of full SI and 434 to 453 mm at full SI. The highest ET values were in the 1992–1993 season.


View this table:
[in this window]
[in a new window]
  		Table 4 Mean squares for the combined analysis of variance for evapotranspiration (ET) and gross water-use efficiency (WUEg) for grain yield (GY) and total dry matter (TDM) during four seasons at Tel Hadya (1992–1996)

 

View this table:
[in this window]
[in a new window]
  		Table 5 Mean evapotranspiration (ET) at different sowing dates, N levels and SI rates during four seasons at Tel Hadya (1992–1996), by N level applied (0–15 g N m-2)

 
Application of 10 or 15 g N m-2 significantly (P < 0.01) increased ET by an average of 44 ± 21 mm with the November sowing and 29 ± 11 mm with the December sowing compared with the treatments without N (Table 5). However, the difference in ET between the 5, 10, and 15 g N m-2 treatments under rainfed conditions and between the 10 and 15 g N m-2 treatments under lower SI was small. A delay in sowing generally decreased ET, except in the 1995–1996 season, when less applied water (150 mm) resulted in a smaller ET for the November sowing than for the December and January sowings.

Cumulative ET was similar under different water treatments at a fixed N rate until anthesis, because the first irrigation was usually applied around the time of anthesis in early April. After anthesis, water use under SI was significantly greater than that under rainfed conditions (Table 6) . The ratio of pre- to post-anthesis ET (ETa/ETpa) was close to 4 for the rainfed treatments, and between 1.2 to 2.4 for the SI treatments at N rates of 10 and 15 g m-2. At the same N rate and sowing date, grain yield and TDM were significantly correlated with ET after anthesis (r = 0.64 to 0.85, P < 0.05, n = 16), but not before anthesis in most of the cases.


View this table:
[in this window]
[in a new window]
  		Table 6 Selected pre-anthesis water use, post-anthesis evapotranspiration (ET), ratio of pre- to post-anthesis ET, grain yield, total dry matter, harvest index and gross water-use efficiency for grain yield and total dry matter of wheat at early sowing during four seasons at Tel Hadya (1992–1996).{dagger}

 
Water Use Efficiency
Water use efficiency was significantly influenced by sowing date, SI, N, and most of the interactions, and varied between seasons (Tables 4 and 7) . In the November sowing, the application of 5 g N m-2 increased WUEr from 0.54 to 0.89 kg m-3 for grain yield. No further increase was achieved with the application of more N under rainfed conditions. A similar trend occurred in the December and January sowings, but at lower WUEr values. The WUEr for TDM ranged from 1.55 to 2.57 kg m-3, with a trend similar to that for grain yield. The WUEr for grain yield and dry matter in November sowing was similar to those of the December sowing. January sowing decreased WUEr from about 0.77 to 0.58 kg m-3 for grain yield and from 2.25 to 1.77 kg m-3 for TDM.


View this table:
[in this window]
[in a new window]
  		Table 7 Rain water-use efficiency and gross water-use efficiency for grain yield (GY) and total dry matter (TDM) averaged over four seasons at Tel Hadya (1992–1996), by level of N applied (0–15 g N m-2)

 
On average, the application of 5, 10, and 15 g N m-2 significantly (P < 0.001) increased WUEg for grain yield from less than 0.6 kg m-3 to between 0.80 and 0.99 kg m-3 in the November and December sowings. The increase in WUEg for grain yield was smaller for the January sowing. A similar trend occurred in WUEg for TDM. As to the variation between seasons, the application of N increased WUEg for grain yield and TDM (P < 0.05), except during the 1993–1994 season, and no difference in mean WUEg for grain yield and TDM occurred between 10 and 15 g N m-2. The 1/3 SI and 2/3 SI treatments significantly (P < 0.01) increased WUEg for grain yield but not for TDM. The highest WUEg for grain yield was achieved at 2/3 SI in the November sowing and 1/3 SI in the December and January sowings. Sowing in November and December consistently resulted in higher WUEg for grain yield and TDM than sowing in January under different N and SI levels. This is clearly associated with the substantial increase in grain yield and TDM with sowing in November and December.

Plotting grain yield and TDM (Table 3) against ET (Table 5) did not give significant linear relations (r2 = 0.45 for grain yield and r2 = 0.46 for TDM) due to the effects of N and sowing date. However, there were good linear relationships between grain yield and TDM with ET for the crop receiving N with 10 and 15 g N m-2 (Fig. 1) , with respective mean WUEg of 1.23 kg m-3 and 3.07 kg m-3 after initial gain of grain and dry matter. Delaying sowing to January decreased WUE for TDM and grain yield.



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1 Relationship of evapotranspiration to grain yield and total dry matter accumulation in wheat. Open circles indicate November sowing; solid squares, December sowing; open triangles, January sowing

 
Water use efficiency for supplemental irrigation (WUESI) ranged from 0.25 to 2.34 kg m-3, depending on the SI level, N rate, and sowing date (Fig. 2) . The WUESI increased with increasing N in the November and December sowings, but there was no further increase beyond 10 g N m-2 for the January sowing. The WUESI also increased when the applied water increased from 1/3 SI to 2/3 SI, but dropped sharply at full irrigation in the November sowing. In the December and January sowings, WUESI decreased after 1/3 SI. Delaying sowing date generally decreased WUESI. The response of WUESI for TDM to the various treatments had a similar trend to that for grain yield. The highest WUESI was achieved at 2/3 SI for the early sowing and at 1/3 SI for the normal and late sowings. Full SI decreased WUESI at lower N rates and had similar WUESI to WUEr at higher N rates.



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 2 Water use efficiency for supplemental irrigation (WUESI) as affected by N rate, SI level, and sowing date

 

   Discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results
 Discussion
REFERENCES
 
In the rainfed farming systems of the Mediterranean region, where water supply is often limited due to low and erratic rain, agronomic practices should aim to utilize the water available for crop growth in an efficient way. Improved production from a limited water supply can result from increasing the total amount of water used by the crop through supplemental irrigation, and improving the efficiency of water use through the adoption of deficit irrigation, N application, and early sowing.

The good correlation between grain yield and post-anthesis ET in our study indicates that grain yield is strongly influenced by the pattern of water use during the course of the season and emphasizes the importance of adequate water supply after anthesis for higher yield and WUE. Supplemental irrigation at or after anthesis not only allows the plant to increase its photosynthesis rate (Passioura, 1976) but, more importantly, gives the plant extra time in which to translocate carbohydrate reserves to the grain (Zhang et al., 1998). The degree to which water supply limits grain yield is indicated by the ratio of ETa/ETpa (2.1–2.4:1). These values are comparable to those found under greenhouse conditions (Passioura, 1983) and under field conditions in South Australia (French and Schultz, 1984). Under SI, water is used more efficiently if combined with adequate N supply. Mean WUEg increased from 0.77 kg m-3 under rainfed conditions to 0.83 to 0.92 kg m-3 under SI. The increase in WUE under SI is associated with the increased leaf area and its effect on the ratio of soil evaporation to crop transpiration (Zhang et al., 1998) and increased root growth and its effect on water extraction (Cooper et al., 1987b).

Early sowing has an advantage in terms of crop grain yield and WUE, particularly where practiced in conjunction with a good N supply and SI. Delayed sowing decreases intercepted solar radiation and reduces the duration of growth (Gregory and Eastham, 1995). Therefore, late-sown crops accumulate less dry matter. Earlier-sown crops have not only increased accumulated dry matter but also reduced water evaporation from the soil surface resulting from an earlier and larger ground cover. Thus, the amount of water transpired by the crop and WUE increase with early sowing. In addition, late-sown crops in this study flowered in early May, a period when the maximum air temperature was above the optimal daily maximum for wheat (23°C) (French and Schultz, 1979; Stapper and Harris, 1989). This heat stress to which the late-sown crops were subjected could have reduced kernel number per ear.

The large variations in grain yield, TDM, and WUEg between seasons can be attributed to seasonal differences in the distribution of rain and air temperature during the grain-filling period. Although rain in the 1992–1993 season was less than the long-term average, the crop must have benefited from the favorable distribution of rain during the growing season. Later rain in the 1994–1995 season did not benefit the crop. In the 1993–1994 season, the crop was adversely affected by a heat wave in April and May, which may have been responsible for the lower crop grain yield. Lower D and air temperature during the grain-filling period (May) in the 1992–1993 season may be responsible for the higher grain yield and TDM and, hence, higher WUEg. The higher D in the 1994–1995 season did not favor grain filling of wheat and reduced grain yield and WUE. Previous studies from 1985 to 1989 (Perrier and Salkini, 1991) also showed that grain yield was much lower with higher D. Combining our data with previous SI experiments (Perrier and Salkini, 1991) conducted at Tel Hadya showed a significantly negative correlation (r = -0.62, n = 52, P < 0.001) between WUE and D during the grain-filling period. In addition, a higher proportion of soil water evaporation in ET in the 1994–1995 and 1995–1996 seasons than in the 1992–1993 and 1993–1994 seasons (Zhang et al., 1998) might also have reduced WUE in these two seasons.

This work shows that the common practice of supplemental irrigation, which aims at satisfying full irrigation requirements, is not the most efficient in water use for the Mediterranean rainfed environment. Applying 1/3 to 2/3 of the SI requirements, during and after anthesis, substantially improves irrigation water productivity. The loss in yield due to deficit irrigation is small compared with the savings in irrigation water. If this is combined with an early sowing and adequate N application, a high yield may be achieved and, in addition, up to 50% of the SI water can be saved for expansion of the area irrigated. Testing these findings at 14 farmer's fields in Syria showed that applying 50% of full SI requirements to rainfed wheat reduced yield by only 10 to 15% (unpublished work by the first author, 1996–1999). When water, not land, is the limiting factor for wheat production, a real potential for higher production exists with these water savings.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the help of Dr. M. Singh in the statistical analysis and Dr. G. Ortiz-Ferrara for variety selection and seed supply. We also thank Drs. J. Ryan and A. Hachum for helpful comments on the draft of the paper. Financial support was provided by the Bundesministerium für Wirtschaftliche Zusammenarbeit (BMZ) through the Restricted Core Program on Water Management in West Asia and North Africa.

Received for publication June 9, 1998.
   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results
 Discussion
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
K.-I. Kim, D. E. Clay, C. G. Carlson, S. A. Clay, and T. Trooien
Do Synergistic Relationships between Nitrogen and Water Influence the Ability of Corn to Use Nitrogen Derived from Fertilizer and Soil?
Agron. J., May 7, 2008; 100(3): 551 - 556.
[Abstract] [Full Text] [PDF]

Home page
 J Exp BotHome page
N. C. Turner
Agronomic options for improving rainfall-use efficiency of crops in dryland farming systems
J. Exp. Bot., November 1, 2004; 55(407): 2413 - 2425.
[Abstract] [Full Text] [PDF]

Home page
 ANN BOT (LOND)Home page
J. L. ARAUS, G. A. SLAFER, M. P. REYNOLDS, and C. ROYO
Plant Breeding and Drought in C3 Cereals: What Should We Breed For?
Ann. Bot., June 15, 2002; 89(7): 925 - 940.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
J. K. F. Roygard, M. M. Alley, and R. Khosla
No-Till Corn Yields and Water Balance in the Mid-Atlantic Coastal Plain
Agron. J., May 1, 2002; 94(3): 612 - 623.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (35)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Agricola
Right arrow Articles by Oweis, T.
Right arrow Articles by Pala, M.
Related Collections
Right arrow Agroclimatology
Right arrow Water Use
Right arrow Wheat

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome