Relationship between carbon isotope discrimination ( ), yield and water use efﬁciency of durum wheat in Northern Syria

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Agricultural Water Management 98 (2011) 1856– 1866

Contents lists available at ScienceDirect

Agricultural Water Management

jo u rn al hom epag e: www.elsev ier .com/ locate /agwat

Relationship between carbon isotope discrimination (�), yield and water useefficiency of durum wheat in Northern Syria

A. Wahbi ∗, A.S.A. ShaabanUniversity of Aleppo, Aleppo, Syria

a r t i c l e i n f o

Article history:Received 28 December 2010Accepted 30 June 2011Available online 10 August 2011

Keywords:Carbon isotope discrimination (�)Durum wheatWater use efficiencyNorthern Syria

a b s t r a c t

Field experiments were conducted over two seasons (2005–06 and 2006–07) using six durum wheatgenotypes varying in grain carbon isotope discrimination (�) but similar in phenology. Experimentalsites at Tel Hadya, Breda, Muslmieh and Yahmoul across rainfall zones in Northern Syria were used. Themain objective of this study is to relate the wheat yields of the different genotypes with � and otherplant traits.

Plant growth was monitored throughout the growing season. � was measured for every plot in thegrain (�G) and the 4th fully expanded leaf (�L). Also, ash and protein contents of grain were analyzed.Soil water content throughout the growing season was measured at some sites to calculate water use.

Yield variability between sites was mostly due to rainfall, with variability between cultivars within asite being due to season. �L was higher, with no stress evident at this stage, compared with �G. �L,ranging from 19.97 to 22.76‰, was lower at the drier sites and higher at the wettest site, which changedto the opposites for �G ranging from 12.88 to 16.62‰. Positive correlations were found between � andyields of grain and biomass. The stronger correlations were at the drier sites, whilst at the wettest sitethere was no significant relationship. �L for the different genotypes was more stable between sites andseasons than �G. Water use efficiency did not differ between genotypes, unlike �. � was also correlatedto grain ash and protein content. This suggests to use these cheap and easily obtainable plant traits assurrogate if � was unavailable.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

The primary challenge in wheat improvement programmes inMediterranean areas is to improve drought tolerance, yield stabilityand water use efficiency (Monneveux and Belhassen, 1996). Wheatbreeders at the International Centre for Agricultural Research inthe Dry Areas (ICARDA) have conducted more than two decades ofresearch to develop, select and test genotypic performance againstdifferent biotic and abiotic stresses for yield stability and grainquality (Nachit and Eloufi, 2004).

Several breeding methods and selection approaches have beenused, for example, individual selection, crossing and backcrossing,stress physiology and molecular markers (SSR, AFLP, RFLP). Geneticmaps for durum wheat are developed to determine the responsiblegenome regions linked to traits of interest (Nachit, 2002). How-ever, this research requires a long period of time, and results aresometimes contradictory due to G × E interactions.

∗ Corresponding author at: Faculty of Agriculture, University of Aleppo, PO Box8047, Aleppo, Syria. Tel.: +963 933 680457; fax: +963 21 2 229184.

E-mail address: [email protected] (A. Wahbi).

A complementary selection strategy is the use of carbon isotopediscrimination (�) technique, which could shorten the period ofthe traditional breeding program (Condon and Hall, 1997). Manyresearchers (Condon et al., 1987; Farquhar et al., 1989; Acevedo,1993; Condon and Richards, 1993; Lopez-Castaneda and Richards,1994; Araus et al., 1998; Voltas et al., 1999) suggested the use ofthis technique as an indirect tool for selection of high transpirationefficiency.

Research at ICARDA aims to improve the water use efficiency(WUE) of crops (Nachit, 1998). In water-limited environments, cropyield is a function of water use, WUE and harvest index (Passioura,1977). WUE or water productivity is defined as aerial biomassyield over water use and is a trait that has been proposed as acriterion for yield improvement under drought (Rebetzke et al.,2002; Condon and Richards, 1992). Breeding for improved WUEhas, however, been difficult due to lack of easy screening method-ology. Direct measurement of WUE or transpiration efficiency (TE)is difficult, but crop productivity (e.g. grain and biomass yields) andmorpho-physiological traits (e.g. canopy temperature and stomatalconductivity) can provide information on TE (Araus et al., 2002).

Farquhar et al. (1982) found that the extent to which C3 plantsdiscriminate against � during carbon assimilation was related totheir WUE. So, it is possible to use � as an index of TE since � and

0378-3774/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.agwat.2011.06.012


A. Wahbi, A.S.A. Shaaban / Agricultural Water Management 98 (2011) 1856– 1866 1857

Table 1Site characteristics.

Site Seasonsa Latitude (◦N) Longitude (◦E) Hasl (m)b Rainfall (mm year−1)c

Tel Hadya (TH) 1st, 2nd 36.01 36.56 284 325Breda (BR) 1st, 2nd 35.56 37.10 300 225Muslmieh (MO) 2nd 36.20 37.13 450 350Yahmoul (YH) 2nd 35.36 37.03 510 420

a 1st, 2005–06; 2nd, 2006–07.b Hasl, height above sea level.c Long-term average (1980–2007 for BR and TH and 1992–2007 for MO and YH).

TE are positively related (Farquhar et al., 1982). Differences in �values between varieties at the one site reflect their different CO2assimilation and transpiration efficiencies. Hence plant breederscan then choose the variety that has high � without the need formany years of testing (Nachit, 2002).

Positive relationships between � and yield were observed inmany studies in Mediterranean environments (Acevedo, 1991;Morgan et al., 1993; Sayre et al., 1995; Condon and Hall, 1997;Nachit, 1998; Merah et al., 2001). This relationship was also foundto be positive in durum wheat varieties at two locations and overseveral seasons in northern Syria (Nachit, 1998; Nachit and Eloufi,2004).

The objectives of the present investigation were (i) to determinethe relationships between � for grain (�G) and 4th fully expandedleaf (�L) and yields of grain and biomass, and (ii) to evaluate surro-gates for the expensive � analysis, as well as their stability acrosstime and space in dry, rainfed Mediterranean-type environment.

2. Materials and methods

2.1. Experimental site and soils

2.1.1. Site characterizationLong-term annual rainfall data show Breda as the driest site,

Yahmoul the wettest and Tel Hadya and Muslmieh as intermediate(Table 1).

At Tel Hadya, the seasonal (from October 1st to September 30th)average maximum temperature ranged between 23.3 and 26.4 ◦Cwhilst the average minimum temperature ranged between 8.7 and11.2 ◦C. Solar radiation ranged between 14.2 and 19 MJ m−2 day−1

and total rainfall was as low as 225 and as high as 517 mm year−1

(Table 2). At Breda, the seasonal average maximum temperatureranged between 23.1 and 26.4 ◦C whilst the minimum temperatureranged between 8.5 and 11.7 ◦C. Solar radiation ranged between17.7 and 24.4 MJ m−2 day−1 and total rainfall was as low as 183 mmand as high as 414 mm year−1 (Table 2). At Muslmieh, the sea-sonal average maximum temperature ranged between 23.1 and26.9 ◦C, whilst the minimum temperature ranged between 8.2 and12.7 ◦C. Solar radiation ranged between 17.8 and 18.8 MJ m−2 day−1

and total rainfall was as low as 133 mm year−1 and as high as460 mm year−1 (Table 3). At Yahmoul, the seasonal average max-imum temperature ranged between 19.2 and 24.6 ◦C whilst theminimum temperature ranged between 6.8 and 13.6 ◦C, Solarradiation ranged between 11.3 and 18.4 MJ m−2 day−1 and totalrainfall was as low as 355 mm year−1 and as high as 740 mm year−1

(Table 3).The hottest months at Tel Hadya were July and August reach-

ing about 37 ◦C, and the coldest month was January reaching about1.3 ◦C. The wettest month was January reaching about 69 mm anddecreased at anthesis and grain filling periods to 33 and 17 mm,respectively. The hottest months at Breda were July and August,reaching about 37 ◦C, and the coldest month was December, reach-ing about 1.8 ◦C. On the other hand, the wettest months were

Table 2Average maximum and minimum air temperature, solar radiation and total rainfall at Breda and Tel Hadya (1980–2007).

Year (October 1 –September 30) Max. temp. (◦C) Min. temp. (◦C) Solar radiation (MJ m−2 day−1) Precipitation (mm)

Breda Tel Hadya Breda Tel Hadya Breda Tel Hadya Breda Tel Hadya

1980–81 25.3 25.0 10.0 10.6 19.0 16.0 293 3631981–82 23.5 24.4 9.2 9.8 19.4 15.4 331 3511982–83 23.1 23.3 8.5 9.3 19.4 14.2 278 3331983–84 25.0 24.6 9.8 10.5 20.3 18.4 204 2441984–85 23.9 24.3 9.4 10.3 19.1 18.2 277 3531985–86 24.8 25.3 10.5 10.7 19.3 18.2 218 2841986–87 24.5 24.9 9.6 9.9 19.5 18.4 245 3471987–88 24.5 24.6 10.8 11.0 19.9 17.7 414 5171988–89 25.6 25.6 9.8 10.1 19.5 19.0 195 2251989–90 24.9 25.0 10.0 9.8 22.4 18.6 183 2331990–91 24.8 25.1 10.7 10.8 21.9 18.1 241 3361991–92 23.3 23.5 8.9 9.1 19.5 17.6 274 3471992–93 23.7 24.5 9.1 9.3 18.4 18.1 272 2841993–94 23.9 25.9 9.9 11.1 18.3 18.3 303 3901994–95 23.8 25.2 10.3 11.0 17.9 17.9 232 3141995–96 23.2 24.7 10.2 10.6 17.8 17.7 365 4111996–97 23.3 24.4 9.5 9.9 18.2 18.2 243 4021997–98 24.8 25.7 11.2 11.0 18.0 17.3 211 4091998–99 25.6 26.4 11.1 11.2 18.6 18.4 198 3321999–00 26.4 26.1 11.1 9.9 18.8 18.6 231 3032000–01 25.7 25.6 11.7 11.1 18.3 17.9 367 4152001–02 24.4 25.1 10.7 10.7 18.8 18.1 340 4102002–03 24.4 24.6 11.0 10.8 18.2 17.8 396 4752003–04 24.7 25.0 10.8 10.9 18.9 17.3 294 4312004–05 24.3 25.0 10.4 10.8 19.0 17.6 268 3272005–06 25.2 25.7 10.0 11.0 17.7 17.9 239 3212006–07 23.4 23.9 9.4 8.7 18.3 17.2 265 313


1858 A. Wahbi, A.S.A. Shaaban / Agricultural Water Management 98 (2011) 1856– 1866

Table 3Average maximum and minimum air temperature, solar radiation and total rainfall at Muslmieh and Yahmoul (1992–2007).

Year (October 1 –September 30) Max. temp. (◦C) Min. temp. (◦C) Solar radiation (MJ m−2 day−1) Precipitation (mm)

Muslmieh Yahmoul Muslmieh Yahmoul Muslmieh Yahmoul Muslmieh Yahmoul

1992–93 23.3 na 8.4 na 18.4 na 248 na1993–94 24.8 19.7 10.1 13.6 18.4 11.3 453 4361994–95 23.9 19.2 10.5 12.5 17.8 12.3 325 4161995–96 23.4 21.5 9.3 11.6 18.0 15.2 460 7411996–97 23.1 21.6 9.1 11.4 17.9 15.4 397 6671997–98 24.2 23.0 10.1 10.9 18.0 16.9 335 5711998–99 25.5 24.2 10.3 11.3 18.6 17.5 230 3821999–00 24.9 24.4 9.4 10.1 18.8 18.4 214 3922000–01 24.8 24.4 10.7 10.6 17.8 18.1 361 3912001–02 26.5 24.0 12.4 9.8 17.9 18.4 226 5342002–03 26.9 24.6 12.7 10.5 18.0 18.3 133 5562003–04 26.9 24.2 12.7 10.6 18.1 18.1 153 4612004–05 na 23.9 na 10.7 na 17.8 na 4072005–06 na na na na na na na na2006–07a 23.9 20.0 8.2 6.8 18.8 16.3 236 355

a November–August, Muslmieh; October–June, Yahmoul; na, not available.

January and February (about 50 mm) and decreased during anthesisand grain filling to 24 and 17 mm, respectively. The hottest monthat Muslmieh was July reaching approximately 37.5 ◦C whilst thecoldest month was February reaching about 1.0 ◦C. On the otherhand, the wettest month was January with about 49 mm of rain-fall and decreased at anthesis and grain filling periods to 31 and23 mm, respectively. The hottest month at Yahmoul was Augustreaching about 35.5 ◦C and coldest month was December reach-ing about 2.1 ◦C. The wettest month was December reaching about90 mm and decreased at anthesis and grain filling periods to 56 and19 mm, respectively.

All soils were alkaline (pH range 7.5–8.5) with large amounts offree carbonate present at Tel Hadya, Breda and Muslmieh (reach-ing about 300 g kg−1, Shaaban, 2009). The Yahmoul soil had lesscarbonate (below 30 g kg−1) and the lowest pH (around 7.5) com-pared with the other sites. All sites were low in total N (below1 g kg−1) and organic matter (about 10 g kg−1), and Olsen-P wasmoderate to low (between 5 and 10 ppm). Soil porosity was opti-mum at Tel Hadya, Breda and Muslmieh and moderate at Yahmoul(clay content was 64%; Shaaban, 2009).

3. Experimental

The experiment was conducted during 2005–06 and 2006–07at Tel Hadya and Breda, whilst in the second season also atMuslmieh and Yahmoul. Six durum wheat genotypes with con-trasting grain-� values but similar phenology were selected.The genotypes, supplied by ICARDA, were Omruf-2 (� = 14.78‰),Brachoua (� = 14.88‰), Korifla (Cham3) (� = 14.95‰), Waha(Cham1) (� = 15.43‰), Massara-1 (� = 15.71‰) and Omrabi-14(� = 15.88‰). The first three genotypes were classed as low � andthe other three as high � according to suppliers.

The experimental design was randomized complete block withthree replicates. A fallow plot was also included in the design.Plot size was between 4 and 32 m2 with row spacing of 17.5 cm.Seeds were planted using an Oyjord planter with a seed rate of

120 kg ha−1. Sowing dates were around the first week of Decem-ber. Optimum agronomic practices were used including cultivation,fertilization following pre-sowing soil analysis, and weed control.Daily weather data was recorded at each site.

3.1. Measurements

3.1.1. Soil moistureSoil water content was monitored for every plot at Tel Hadya

(2005–06 and 2006–07) using the Diviner 2000, a capacitance sen-sor manufactured by Sentek (Sentek sensor technologies, Stepney,Australia). PVC access tubes were manually installed to a depthof about 160 cm and measurements were taken throughout theseason at 10 cm intervals. The sensor was calibrated at the end ofthe season when the soil was very dry by measuring with Divinerand taking soil samples to determine water contents. Then wetthe soil profile and measure by Diviner and taking soil samples,then wait for few of days till field capacity is reached. Repeat themeasurements of Diviner and taking soil samples (to cover the fullrange of soil wetness). Calibration curve was derived from thesedata and used in the Diviner measurements. The water used perplot (evapotranspiration) was calculated from the sensor measure-ments assuming that no runoff or drainage below the measureddepth of 160 cm had occurred.

Water use efficiency (WUE) for biomass and grain were deter-mined for each genotype by dividing the biomass or grain yield bythe amount of evapotranspiration.

3.2. Plant measurements

Plant samples were taken from every plot to monitor plantgrowth throughout the season. The samples taken according to site,season, plant developmental stage and area sampled are shown inTable 2. Plants were cut at ground level, dried in an oven at 70 ◦Cfor 48 h and weighed for dry matter.

Table 4Plot area (m2) sampled at different growth stages at four sites between 2005–06 and 2006–07.

Site Season 3–4 leaf Tillering Stem elong. Anthesis Maturity

Breda 2005–06 0.4 0.4 0.2 0.2 1.02006–07 0.8 0.2 – 0.2 1.0

Tel Hadya 2005–06 0.8 0.6 0.2 0.2 1.02006–07 0.8 0.2 – 0.2 1.0

Muslmieh 2006–07 0.7 0.175 – 0.175 0.875Yahmoul 2006–07 0.8 0.2 – 0.2 0.8



Fig. 1. Dry matter production (kg ha−1) for five growth stages expressed as thermal unites, TU (data in parentheses are days after sowing), DAS for the six genotypes at twolocations (BR, Breda; TH, Tel Hadya) in 2005–06.

The 4th fully expanded leaf was sampled for each replicate forthe six genotypes in 2005–06 and for two replicates in 2006–07,and grain was taken from every plot. Leaf and grain samples weredried at 70 ◦C and finely ground for � analysis at the InternationalAtomic Energy Agency (IAEA) Laboratories, Seibersdorf, Austria. �was calculated according to the procedure described by Farquharet al. (1989).

Spikes (10 each) were sampled from anthesis till maturity every3–4 days to monitor the grain accumulation during grain filling.

Grain protein and ash content were determined, at ICARDA labo-ratory, using near infrared spectroscopy (Foss-NIR Systems Europe,Raamsdonksveer, Holland) at all sites in 2005–06 and 2006–07(Table 4).

Results were statistically analyzed using Genstat V 7.0 and SPSS15.0 software packages and stability analysis using GGE biplot(www.ggebiplot.com) software packages.

4. Results

4.1. Seasonal dry matter accumulation

4.1.1. 2005–06The season started wet between December and March and then

turned dry for April and May only 25 and 1 mm, respectively, atTel Hadya and 26 and 2 mm, respectively at Breda. Total rainfallduring the growing season was 135 mm at Breda and 226 mm atTel Hadya. Biomass production throughout the season was verysimilar among genotypes with no significant differences (except atthe 3–4 leaf stage at Tel Hadya and at physiological maturity at TelHadya (P < 0.05; Fig. 1).

At Breda, vigorous plants were evident at the beginning ofthe season, even better than that of Tel Hadya (at the 3–4 leafstage; Fig. 1), but thereafter, the crop was badly affected by thedrought and hot spell, and produced 0.3–0.7 t ha−1 grain by theend of the season. It is interesting to note that at Breda, the above-ground biomass yield reached its maximum at stem elongation,stayed almost constant till anthesis and then decreased sharply

and dramatically to physiological maturity (Fig. 1). Such obser-vation was also reported by El Farhan (2000) for wheat plantsgrown in Syria under heat stress. Grain yield at Tel Hadya wasbetween 1.9 and 2.2 t ha−1 with no significant differences betweengenotypes.

4.1.2. 2006–07This season started with little rain and was wet in May, the oppo-

site to that of the previous season (2005–06). Total rainfall duringgrowing season was: 150, 188, 195 and 266 mm at Breda, Muslmieh,Tel Hadya, and Yahmoul, respectively. A significant amount of rainhad fallen in May reaching 19, 29, 42, and 53 mm at Muslmieh,Yahmoul, Breda, and Tel Hadya, respectively. However, the cropssuffered from frost damage (2, 11, 32 and 66 days where air temper-ature drop below zero at Breda, Tel Hadya, Yahmoul and Muslmieh,respectively) as well as relatively high air temperature at the grainfilling period, causing low yields (Fig. 2). The extreme weather con-ditions occurred mainly at Muslmieh, where total biomass was lessthan 4 t ha−1, which was similar to that at Breda despite the wet-ter conditions at Muslmieh. Also, the crop at Muslmieh reachedmaturity late compared with Breda (growth periods of 186 and123 days, respectively). An outstanding yield was recorded at Yah-moul (266 mm of rain, and a growth period of 180 days) with amarked increase in yield from anthesis to maturity (Fig. 2). Similarresults of the increase in yield, in wet area, were found for the wheatcrop grown in Syria by Al-Ouda (1997). Unlike the other three siteswhere a marked decrease occurred during this period (Fig. 2). Therewere no significant differences in biomass and grain yield betweenthe six genotypes.

4.2. Dry matter accumulation during grain filling

The grain filling periods in 2006–07 were 16, 21, 22 and 28 daysat Breda, Muslmieh, Tel Hadya and Yahmoul, respectively. Massara-1 and Waha accumulated the highest dry matter production duringthe grain filling period at Tel Hadya (highest slope of the relation;Table 5), whilst Massara-1 continued to be superior at Breda and



Fig. 2. Dry matter production (kg ha−1) for four growth stages for the six genotypes at four locations (YH, Yahmoul; TH, Tel Hadya; BR, Breda; MO, Muslmieh) in 2006–07.

Table 5Linear relationships between dry matter production (Y, kg ha−1) at grain-filling and days after sowing (x) for six genotypes at four locations in 2006–07.

Genotype Tel Hadya Breda Muslmieh Yahmoul

Omruf-2 57.45x − 5197 (0.98) 24.81x − 1283 (0.76) 65.71x − 9081 (0.97) 154.6x − 19920 (0.96)Brachoua 50.68x − 4358 (0.88) 32.67x − 2381 (0.84) 66.79x − 9273 (0.99) 140.5x − 18460 (0.97)Korifla 57.46x − 5408 (0.87) 32.9x − 2228 (0.66) 113.4x − 16280 (0.94) 142.9x − 18710 (0.98)Waha 81.03x − 7137 (0.99) 29.73x − 2061 (0.93) 85.54x − 12130 (0.99) 130.7x − 16330 (0.99)Massara-1 88.4x − 8247 (0.95) 46.83x − 3632 (0.63) 84.43x − 12190 (0.97) 163.6x − 21330 (0.97)Omrabi-14 74.06x − 7387 (0.88) 23.67x − 1308 (0.91) 76.34x − 10990 (0.88) 140.7x − 18240 (0.99)

Data in parentheses are R2 of the linear relation.

Fig. 3. Relationships between grain yield (kg ha−1) and �G (‰) for six durum wheat genotypes at higher rainfall sites of Tel Hadya in 2005–06 and Yahmoul in 2006–07.



Table 6Evapotranspiration (mm) for six durum wheat genotypes from sowing till stages listed at Tel Hadya in 2005–06 (1) and 2006–07 (2).

Genotypes � 3–4 leaf Tillering Stem elongation Anthesis Maturity

1 1 2 1 1 2 1 2

Omruf-2 Low 111 122 144 189 285 164 288 203Brachoua 109 119 148 182 271 183 277 228Korifla 105 115 151 184 276 169 283 207Waha High 109 124 148 188 290 167 296 213Massara-1 109 117 152 179 286 182 289 223Omrabi-14 108 119 144 190 280 165 287 202No crop–Fallow 106 115 116 140 144 142 185 206SE (±) 10.2ns 12.6ns 14.7ns 13.8* 18.6*** 15.2ns 15.3*** 17.2ns

ns, not significant; SE, standard error.* Significant at P < 0.05.

*** Significant at P < 0.001.

Table 7Water use efficiency (kg ha−1 mm−1) for six durum wheat genotypes for biomass production till stages listed and for grain yield at Tel Hadya in 2005–06 (1) and 2006–07 (2).

Genotypes 3–4 leaf Tillering Stem elongation Anthesis Biomas yield Grain yield

1 1 2 1 1 2 1 2 1 2

Omruf-2 2.1 14.1 24.8 51.2 41.8 27.8 33.8 26.0 7.1 1.3Brachoua 2.0 16.1 10.0 46.3 41.3 28.8 36.3 21.0 7.0 3.4Korifla 2.3 16.1 22.3 51.0 38.8 32.6 34.1 24.7 7.9 2.1Waha 1.6 14.3 21.6 40.3 40.1 34.6 31.2 26.4 7.0 3.3Massara-1 1.4 12.7 15.8 42.0 40.0 30.3 36.2 24.4 7.2 3.4Omrabi-14 1.4 13.0 26.1 38.5 41.6 31.5 33.2 23.1 6.4 1.6SE (±) 0.4ns 2.4ns 3.6** 5.7ns 5.2ns 7.7ns 2.9ns 2.5ns 0.9ns 0.5**

ns, not significant; SE, standard error.** Significant at P < 0.01.

Yahmoul. Korifla accumulated the highest dry matter at Muslmieh.The average of four sites for daily dry matter accumulation was 72.7,75.6, 78.7, 81.8, 86.7 and 95.8 kg ha−1 day−1 for Brachoua, Omruf-2, Omrabi-14, Waha, Korifla and Massara-1, respectively. Whereasthe site average were 31.8, 68.2, 82.0 and 145.5 kg ha−1 day−1 forBreda, Tel Hadya, Muslmieh and Yahmoul, respectively.

4.3. Water use and water use efficiency

The total water used (evapotranspiration) was similar betweenthe different genotypes and varied at maturity between 277 and296 mm throughout the 2005–06 growing season at Tel Hadya(Table 6). However, significant differences were recorded betweenthe genotypes and fallow (bare soil) at different plant stages

(Table 6). Moreover, the high-� genotypes used more water bythe end of the season (anthesis, and physiological maturity) com-pared with the low-� genotypes (not significant, Table 6). WUEshowed that, the higher-� genotypes, the lower WUE (althoughnot significant, Table 7).

Unlike 2005–06, the genotypes and fallow did not differ signifi-cantly in total water used at Tel Hadya during the 2006–07 season(varied between 202 and 228 mm; Table 6). However, in this sea-son, genotypes showed significant differences in WUE at tilleringand maturity for grain yield at Tel Hadya in 2006–07 (Table 7).And unlike 2005–06, the low-� genotypes had a significantly (atP < 0.01; Table 7) lower WUE for grain yield compared with thehigh-� genotypes. Similar results were reported in a field study inSyria for wheat crop by Eberbach and Pala (2005).

Table 8Fourth fully expanded leaf-� (�L) and grain-� (�G) (‰) of six durum wheat genotypes for two sites in 2005–06 and four sites in 2006–07.

Genotypes 2005–06 2006–07 Average

Tel Hadya Breda Tel Hadya Breda Muslmieh Yahmoul

�L (‰)Omruf-2 21.95 21.05 21.45 20.71 19.97 22.00 21.19Brachoua 22.36 21.29 21.12 21.10 20.43 21.83 21.36Korifla 22.52 21.55 21.71 21.13 20.48 22.19 21.60Waha 22.51 21.57 22.08 21.08 20.49 22.28 21.67Massara-1 22.39 21.63 21.96 20.95 20.74 21.28 21.49Omrabi-14 22.76 21.83 22.20 21.38 20.85 22.57 21.93SE (±) 0.07*** 0.07*** 0.27ns 0.14ns 0.30ns 0.55ns

�G (‰)Omruf-2 14.78 15.36 15.35 13.72 14.08 12.88 14.36Brachoua 14.88 15.33 15.70 14.02 14.78 13.63 14.72Korifla 14.95 15.72 15.68 14.06 14.83 13.69 14.82Waha 15.43 15.80 15.81 14.21 14.88 13.75 14.98Massara-1 15.71 16.13 15.62 14.13 15.21 13.70 15.08Omrabi-14 15.88 16.26 15.71 14.08 15.17 13.71 15.14SE (±) 0.12** 0.07*** 0.08** 0.19*** 0.16*** 0.17*


** Significant at P < 0.01.*** Significant at P < 0.001.



Table 9Entry means and Shukla’s stability index for the 4th fully expanded leaf-� (�L) and grain-� (�G).

� Parameter Omruf-2 Brachoua Korifla Waha Massara-1 Omrabi-14

�L Mean −0.350 −0.184 0.058 0.130 −0.047 0.393SBD 0.163 0.218 0.070 0.121 0.336 0.079

�G Mean −0.398 −0.153 −0.011 0.094 0.165 0.303SBD 0.216 0.235 0.163 0.165 0.193 0.199

SBD, stability deviation, calculated as the square root of the GE mean square associated with each entry, and a larger value denotes more GE interaction and less stability.

4.4. Carbon isotope discrimination

4.4.1. 4th fully expanded leafThere were significant differences in �L between genotypes for

the two sites in 2005–06, ranging from 19.97 to 22.76‰, but not in2006–07 (Table 8). Korifla and Waha were intermediate �L andOmrabi-14 the highest (Table 8). This variability was consistentamong seasons and sites as shown low stability deviation in theShukla’s Stability Index (Table 9), which refer to stable numbers(numbers and ranking) for the different sites and seasons.

4.5. Grain

There were significant differences in �G between genotypes,ranging between 12.88 and 16.62‰, at all sites and seasons, exceptfor the wettest site, Yahmoul in 2006–07 (Table 8). In contrast to�L, �G was highest at the driest sites (e.g. Breda) and lowest at thewettest sites (e.g. Yahmoul). Consistent with �L, Korifla and Wahawere intermediate �G and Omrabi-14 the highest (Table 8). Thisvariability was consistent among seasons and sites as shown withthe low stability deviation in the Shukla’s Stability Index (Table 9).

4.6. Ash and protein content

Only at Breda there were significant differences in the grainash content between the different genotypes for the two seasons2005–06 and 2006–07 (Table 10). At Breda 2005–06 season, thehigher-� genotypes, the higher its grain ash content (significantat P < 0.05; Table 10). Breda is the driest among the four sites.For the other sites, similar trends were observed but with no sig-nificant differences between genotypes. There was a tendency ofhigher ash content for the drier site compared with wetter site(Table 10).

For the grain protein content, significant differences betweengenotypes were observed at Breda for the first season (2005–06)and at Breda, Tel Hadya and Muslmieh for the 2006–07 season(Table 10). With a tendency of higher grain protein content for thedrier site compared with wetter site (similar to ash grain content)and, to a less extend, higher-� genotypes compared with lower-�genotypes.

So, both ash and protein (grain) content related positively with� and dry site (e.g. Breda), and Waha (Cham1) had the highest ashand protein (grain) content (traits that are closely related to droughttolerance (Nachit et al., 1995) compared with the other locationsand genotypes.

4.7. Relationships between plant traits

4.7.1. Grain yield versus �GAt the wetter sites Tel Hadya in 2005–06 and Yahmoul in

2006–07, with seasonal rainfall of 226 and 266 mm, respectively,relationships between grain yield and �G were, both, negative andpositive (Fig. 3). Similar negative relations were found for the Aus-tralian wheat cultivars (e.g. Farquhar and Richards, 1984; Ehdaieet al., 1991). However, a positive relationships between grain yieldand �G at the moderate rainfall sites at Tel Hadya in 2004–05 (afterShaaban, 2009) and 2006–07, where seasonal rainfall was 164 and195 mm, respectively, and at Muslmieh in 2006–07 where seasonalrainfall was 188 mm (Fig. 4) except for Omruf-2 where a negativerelation with low R2 of 0.17 was evident. Similar trends for thedifferent genotypes (quality), but differed in quantity (slope of thelinear equation; means grain production per ha per unit of �G) wasevident, with R2 differed between the genotypes and there was atendency for a higher R2 and slope (indication of production perunit of �G) of the linear relationship for the high �G genotypes(Fig. 4). So, the higher production genotypes showed a reasonably

Table 10Grain ash and protein content (%) for six durum wheat genotypes at two sites in 2005–06 and four sites in 2006–07.

Genotypes 2005–06 2006–07 Average

Tel Hadya Breda Tel Hadya Breda Muslmieh Yahmoul

Grain ash content (%)Omruf-2 3.13 3.21 3.21 3.23 3.10 3.10 3.16Brachoua 3.08 3.11 3.20 3.16 3.06 3.11 3.12Korifla 3.10 3.20 3.13 3.16 3.12 3.14 3.14Waha 3.18 3.29 3.21 3.28 3.09 3.10 3.19Massara-1 3.13 3.32 3.15 3.12 3.07 3.12 3.15Omrabi-14 3.13 3.34 3.23 3.25 3.10 3.10 3.19SE (±) 0.050ns 0.029** 0.030ns 0.040* 0.04ns 0.028ns

Grain protein content (%)Omruf-2 14.3 18.1 16.3 18.5 14.1 12.8 15.7Brachoua 13.4 17.5 15.4 17.2 12.5 13.0 14.8Korifla 13.7 18.1 16.2 17.5 12.1 13.2 15.1Waha 15.1 19.1 18.2 18.9 13.1 12.3 16.1Massara-1 14.0 18.7 15.7 17.8 12.7 12.2 15.2Omrabi-14 14.2 19.2 17.3 18.9 12.5 12.3 15.7SE (±) 0.69ns 0.26** 0.20*** 0.45* 0.45* 0.92ns


** Significant at P < 0.01.*** Significant at P < 0.001.



Fig. 4. Relationships between grain yield (kg ha−1) and �G (‰) for six durum wheat genotypes at moderate rainfall sites of Tel Hadya in 2004–05 and 2006–07, and Muslmiehin 2006–07.

close data to the fitting linear relation (high R2), so it is reasonablystable genotypes at different weather condition.

Results at the drier site Breda, with 135 and 150 mm of sea-sonal rainfall for the 2005–06 and 2006–07 seasons, respectively;were slightly different. One genotype (Brachoua) showed a nega-tive slope of the relation between grain yield and �G, but, indeedvery low R2 (0.09; Fig. 5). Otherwise, positive relationships wereevident with a tendency for higher slopes of the linear relationshipfor the high �G genotypes (Fig. 5).

So, relationships between �G and grain yields for individualsites were positive in the dry sites, but were more diffuse in the

wet sites or in high grain yield sites (Fig. 6). When all data werecombined, the relationship was positive with R2 of 0.49 (Fig. 7).

4.7.2. Biomass yield versus �GA positive linear relationship was found between biomass yield

and �G for the combined genotypes, sites and seasons (Fig. 7).Although, the R2 was small (0.53) this is mainly due to data scat-tered below the line for the site Muslmieh (it was very low yieldproduction and with still high �G) and above the line for the siteYahmoul (which had very high yield and high �G). Similar resultswere found in Syria by (Nachit, 1998; Nachit and Eloufi, 2004).

Fig. 5. Relationships between grain yield (kg ha−1) and grain-� (‰) for six durum wheat genotypes at Breda for two seasons (2005–06 and 2006–07).



Fig. 6. Relationships between grain yield (kg ha−1) and �G (‰) for individual sites (YH, Yahmoul; TH, Tel Hadya; BR, Breda; MO, Muslmieh) and seasons (04–05, 2004–2005;05–06, 2005–2006; 06–07, 2006–2007).

4.7.3. Overall relationships between �L or �G and grain orbiomass yields

Highly positive relationships were found between �G or �L andgrain or biomass yields (Fig. 8). Although, the relations seems to bebetter for �G compared with �L (higher in R2) and also better inbiological yield compared with grain yield (higher R2, Fig. 8).

5. Discussion

The relation between �L or �G and grain or biomass yield wasexamined and showed consistently positive relations. In Syria, pos-itive relation was found between yield and �G in different drylocations (Nachit, 1998; Nachit and Eloufi, 2004). Misra et al. (2006)

reported positive relation between yield of wheat and �G in the hotenvironment of Peninsular India. However, in this study the rela-tionship was very clear and positive at dry sites compared withwet sites (Figs. 3–5). Moreover, and due to severe weather condi-tion of frost that occurred at some sites (Muslmieh 2006–07; frostof 66 days during the growing season), �G decreased comparedwith other site with similar rainfall (Tel Hadya; Table 8). On theother hand, Yahmoul (2006–07) that had the highest precipitationamong the sites, showed a very low �G compared with the othersites (Table 8).

Different hypothesis can explain the association between �Gand grain yield under terminal drought condition. First, high�G and grain yield could reflect high-� values at an early

Fig. 7. Relationship between grain yield (GY), biomass yield (BY) (kg ha−1) and �G (‰) for all durum wheat genotypes across sites and seasons.



Fig. 8. Relationships between biomass yield (BY), grain yield (GY) (kg ha−1) and �G and �L (‰) across sites and seasons.

stage of growth. In this case � may reflect faster growth ratethrough crop development (Condon and Richards, 1993). Secondly,high �G could characterize genotypes with high �L and thatmore dependent pre-anthesis vegetative reverses for grain filling.Under post-anthesis water stress photosynthesis is generally moreaffected than translocation and plants mainly use assimilates frompre-anthesis reserves for grain filling (Loss and Siddique, 1994). Thecontribution of pre-anthesis assimilates to grain filling may varyfrom 10 to 80% of final grain weight (Palta and Fillery, 1995). �Gis influenced by the proportion of reserves used for grain filling. Itmay be higher if a large proportion of starch originates from stemreserves that were accumulated when plant were less stressed andhad high � values. Thirdly, high �G can be the consequences of ear-liness. Finally, relationship between �G could reflect a better abilityto maintain stomata more open after anthesis (Morgan et al., 1993).

The difference between �G and �L was high, taking intoaccount, that no stress occurred at early season, so, �L will show analmost full capacity but at the end of the season, grain showed thesumming up of all the stress occurred during the season (water,heat, etc.), so, �G was not as stable (and lower) measurementscompared with �L among seasons and sites.

Unlike �G or �L, WUE seems not to be different among thegenotypes, this is mainly due to the high temperature and so, allthe water will be evaporated from the soil (if not used by the plantin transpiration).

Shaaban (2009) showed in stepwise regression for all the mea-sured traits that �G correlated highly (P < 0.001) with the grainyield, protein grain content, stomatal resistance and �L. Shaabanalso showed in stepwise regression analysis that �L correlated sig-nificantly (P < 0.05) with straw yield, spikelet number per spike,peduncle length, thousand grain weight and ash grain content. So,from the above results, it could be concluded that grain protein andash contents could be an alternative surrogates of �G or �L, sincethey are very cheap and easy to analyze.

6. Conclusions

�G values varied between 12.88 and 16.62‰ and were higherfor the �L and ranged between 19.97 and 22.76‰. Significant geno-

typic differences were evident for �G in all the locations exceptYahmoul (wettest). Significant genotypic differences were foundfor �L at the two locations (2005–06), but, was no significant geno-typic differences at the four locations (2006–07). Higher variationwas apparent in the �G values compared with variation of the �L,with a similar ranking order between genotypes for �G and �L. Nosignificant G*E interactions were evident for two genotypes, Waha(Cham1) and Korifla (Cham3) for almost all of the measured traits.Higher and positive correlations were found between grain yieldand �G in dry locations compared to moderately wet locations.Significant correlations were evident between �G and grain yield,grain protein content, stomatal resistance and �L. �L was closelyassociated with biomass yield, spiklet number spike−1, pedunclelength, 1000-grain weight and grain ash content. The results ofgrain ash and protein content were associated with results of �G.The genotypes with high-� were high in grain ash and proteincontent. Also, the driest site (Breda) had a high ash and protein con-tent. In all trials Waha (Cham1) was the higher in ash and proteincontent.

Acknowledgements

This work was partially financed by the International AtomicEnergy Agency (IAEA), Vienna, Austria (Research contract SYR13071). Authors acknowledged the editing of the IAEA members.Special appreciation is given to International Centre for AgriculturalResearch in the Dry Areas for providing the durum wheat varietiesand land for the experiment.

References

Acevedo, E., 1991. Improvement of winter cereal crops in Mediterranean envi-ronments, use of yield, morphological and physiological traits. In: Acevedo,E., Conesa, A.P., Monneveux, P., Srivastava, J.P. (Eds.), In Physiology Breedingof Winter Cereals for Stressed Mediterranean Environments. INRA, Paris, pp.273–305.

Acevedo, E., 1993. Potential of carbon isotope discrimination as a selection cri-terion in barley breeding. In: Ehleringer, J.R., Hall, A.E., Farquhar, G.D. (Eds.),Stable Isotopes and Plant Carbon–Water Relations. Academic Press, New York,pp. 399–417.



Al-Ouda, A.S.H., 1997. Assessment the importance of induction and its nature inimproving the tolerance of some ACSAD durum and bread wheat landraces tosalinity stress. Damascus Univ. J. Agric. Sci. 23 (2), 15–30 (in Arabic, abstract inEnglish).

Araus, J.L., Amaro, t., Casadesus, J., Asbati, A., Nachit, M.M., 1998. Relationshipbetween ash content, carbon isotope discrimination and yield in durum wheat.Aust. J. Plant Physiol. 25, 835–842.

Araus, J.L., Slafer, C.A., Reynolds, M.P., Royo, C., 2002. Plant breeding and drought inC3 cereals: what should we breed for? Ann. Bot. 89, 925–940.

Condon, A.G., Hall, A.E., 1997. Adaptation to diverse environments: variation inwater-use efficiency within crop species. In: Jackson, L.E. (Ed.), Ecology in Agri-culture. Academic Press, San Diego, pp. 79–116.

Condon, A.G., Richards, R.A., 1992. Broad sense heritability and geno-type × environment interaction for carbon isotope discrimination in field-grownwheat. Aust. J. Agric. Res. 43, 921–934.

Condon, A.G., Richards, R.A., 1993. Exploiting genetic variation in transpiration effi-ciency in wheat: an agronomic view. In: Ehleringer, J.R., Hall, A.E., Farquhar, G.D.(Eds.), Stable Isotopes and Plant Carbon–Water Relations. Academic Press, NewYork, pp. 435–450.

Condon, A.G., Richards, R.A., Farquhar, G.D., 1987. Carbon isotope discrimination ispositively correlated with grain yield and dry matter production in field-grownwheat. Crop Sci. 27, 996–1001.

Eberbach, P., Pala, M., 2005. Crop row spacing and its influence on the partitioningof evapotranspiration by winter grown wheat in Northern Syria. Plant Soil 268,195–208.

Ehdaie, B., Hall, A.E., Farquhar, G.D., Nguyen, H.T., Waines, J.G., 1991. Water-useefficiency and carbon isotope discrimination in wheat. Crop Sci. 31, 1282–1288.

El Farhan, A.S., 2000. Study of Heat Stress on Wheat Growth, Development andProduction. 40th Science Week, 3rd book, 1st volume. Agricultural ScienceResearches Studies, University of Tishreen, Lattakia, Syria, 4–9 December, Publi-cations of Supreme Council for Science, 307–331 (in Arabic, abstract in English).

Farquhar, G.D., Richards, R.A., 1984. Isotopic composition of plant carbon correlateswith water-use efficiency of wheat genotypes. Aust. J. Plant Physiol. 11, 539–552.

Farquhar, G.D., Ehleringer, J.R., Hubick, K.T., 1989. Carbon isotope discrimination andphotosynthesis. Ann. Rev. Plant. Physiol. Plant. Mol. Biol. 40, 503–537.

Farquhar, G.D., O’Leary, M.H., Berry, J.A., 1982. On the relationship between car-bon isotope discrimination and the intercellular carbon dioxide concentrationin leaves. Aust. J. Plant Physiol. 9, 121–137.

Lopez-Castaneda, C., Richards, R.A., 1994. Variation in temperate cereals in rain-fedenvironments III. Water use and water-use efficiency. Field Crops Res. 39, 85–98.

Loss, S.P., Siddique, K.H.M., 1994. Morphological and physiological traits associ-ated with wheat yield increases in Mediterranean environments. Adv. Agron.52, 229–276.

Merah, O., Deleens, E., Souyris, I., Nachit, M., Monneveux, P., 2001. Stability of carbonisotope discrimination and grain yield in durum wheat. Crop. Sci. 41, 677–681.

Misra, S.C., Rao, V.S., Sheshshayee, M.S., Serraj, R., Monnevaux, P., 2006. Relationshipbetween carbon isotope discrimination, ash content and grain yield in wheat inthe Peninsular zone of India. J. Agron. Crop Sci. 192, 352–362.

Monneveux, P., Belhassen, E., 1996. The diversity of drought adaptation in the wild.Plant Growth Regul. 20, 85–92.

Morgan, J.A., LeCain, D.R., McCaig, T.N., Quick, J.S., 1993. Gas exchange, carbon isotopediscrimination and productivity in winter wheat. Crop Sci. 33, 178–186.

Nachit, M.M., Baum, M., Impiglia, A., Ketata, H., 1995. Studies on some grain qual-ity traits in durum wheat grown in Mediterranean environments. In: Di Fonzo,N., Kaan, F., Nachit, M. (Eds.), Proceedings of the seminar on a “Durum wheatimprovement in Mediterranean region. Options Mediterraneens. Serie A., Sem-inaires Mediterraneens, 22 Zaragoza, Spain, 181–187.

Nachit, M.M., Eloufi, I., 2004. Durum wheat adaptation in the Mediterranean dry-land: breeding, stress physiology, and molecular markers. Crop Science Societyof America and American Society of Agronomy, 677 S. Segoe Rd., Madison, WI53711, USA. Challenges and Strategies for Dryland Agriculture. CSSA SpecialPublication no. 32, 203–218.

Nachit, M.M., 2002. Breeding for improved resistance to drought in durum wheat.Caravan 17, 14–15.

Nachit, M.M., 1998. Association of grain yield in dryland and carbon isotope dis-crimination with molecular markers in durum (Triticum turgidum L. var durum)wheat. In: Proc. 9th Int. Wheat Genetics Symp., Saskatoon Canada, pp. 218–223.

Palta, J., Fillery, I.R.P., 1995. N application increase pre-anthesis contribution ofdry matter to grain yield in wheat grown in duplex soil. Aus. J. Agric. Res. 46,507–518.

Passioura, J.B., 1977. Grain yield, harvest index, and water use of wheat. J. Aust. Inst.Agric. Sci. 43, 117–120.

Rebetzke, G.J., Condon, A.G., Richards, R.A., Farquhar, G.J., 2002. Selection for reducedcarbon-isotope discrimination increases aerial biomass and grain yield of rain-fed bread wheat. Crop Sci. 42, 739–745.

Sayre, K.D., Acevedo, E., Austin, R.B., 1995. Carbon isotope discrimination and grainyield for three bread wheat germplasm groups grown at different levels of waterstress. Field Crops Res. 41, 45–54.

Shaaban, A.S.A., 2009. Study of Durum Wheat Cultivars: Production, Water UseEfficiency, Carbon Isotope Discrimination and Simulation Models. Ph.D. thesis,University of Aleppo, Syria, 296 pp., (in Arabic, abstract in English).

Voltas, J., Romagosa, I., Lafarga, A., Armesto, A.P., Sombrero, A., Araus, J.L., 1999.Genotype by environment interaction for grain yield and carbon isotope dis-crimination of barley in Mediterranean Spain. Aust. J. Agric. Res. 50, 1263–1271,www.ggebiplot.com.

Relationship between carbon isotope discrimination ( ), yield and water use efﬁciency of durum wheat in Northern Syria

Documents