Barley (Hordeum vulgare L.) is one of the most important cereals worldwide and is being increasingly grown in many areas of the world. However, there is limited comparative research of the different plant based methods for N assessing N status of the crop such Nitrogen Nutrition Index (NNI) and chlorophyll meter (CM) readings and its relationship with Nitrogen Use Efficiency (NUE) on barley. A two-year field study was therefore conducted with the objective to determine the effect of N fertilization (0, 60, and 120 kg ha-1) on CM readings, NNI and NUE and its components of four barley cultivars. CM readings and RCM readings were affected by the N treatment and were higher at both N levels compared with the control. N nutrition index varied from 0.75 to 1.03 across years, growth stage and cultivar and was affected by the fertilization level. NUE was higher at the control compared with the two N levels and was correlated with grain yield and negatively correlated with N shoot concentration, NNI, and CM readings. This study provides new information about the effect of N application on chlorophyll meter readings, relative chlorophyll meter readings, NNI, and NUE of barley that can be used for obtaining higher grain yield.
Get your grade
or your money back
using our Essay Writing Service!
Barley (Hordeum vulgare L.) is one of the main cereals that is grown in many areas of the world reaching 54 million hectares and total production of 150 million Mg (FAO 2011). However, there is a lack of information about the use of diagnostic tools for nutrient requirements for barley and especially on N which is one of the most important nutrients required for higher grain yield. Efficient use of applied N is also important in barley for higher production and in order to maximize producer's economic returns and maintain soil and water quality. Several diagnostic tools have been developed in order to determine N deficiency which is used to improve N management and decrease the risk of N loss to ground and surface waters (Lemaire et al. 2008; Fageria and Baligar 2005).
The plant-based diagnostic methods that were developed such as chlorophyll meters (CM) are quite useful since they provide a valuable estimation of the N status of the crop (Lemaire et al. 2008; Lemaire and Gastal 2009). However, CM readings remain one of the most popular approaches and have been proven to be effective as a rapid diagnostic method to determine the N status of many crops, including spring wheat (Triticum aestivum L.) (Follett et al. 1992; Vidal et al. 1999; Arregui et al. 2006Â ; Ziadi et al. 2010), rice (Oryza sativa L.) (Turner and Jund 1991; Peng et al. 1993; Ladha et al. 1998), safflower (Carthamus tinctorius L.) (Dordas and Sioulas 2008), and corn (Zea mays L.) (Piekielek and Fox 1992; Dordas et al. 2008; Ziadi et al. 2008). The results were based on the relationship between CM readings and relative grain yield (RY) (Piekielek and Fox, 1992; Blackmer and Schepers, 1994; Fox et al., 2001). However, this relationship can vary with plant development and is lower at the vegetative stage (Blackmer and Schepers 1995; Waskom et al. 1996; Bullock and Anderson 1998) and higher at the later developmental stages (Blackmer and Schepers 1995; Smeal and Zhang 1994; Piekielek et al. 1995; Waskom et al. 1996; Bullock and Anderson 1998). Also CM readings have the disadvantages that they depend on cultivar, management techniques, site characteristics, disease or insect damage, plant density, and other nutrient deficiencies (Blackmer and Schepers 1995; Piekielek et al. 1995; Waskom et al., 1996) and this created the need to find a more stable diagnostic tool. Thus, the relative CM readings (RCM) have been proposed and was found to account for the influence of these factors (Ziadi et al. 2008; Debaeke et al. 2006; Prost and Jeuffrey 2007). The RCM readings are calculated by dividing the readings from the test area by the readings from a saturated plot that has received a high N rate.
Another approach has been proposed which is used to determine the level of plant N nutrition is the N nutrition index (NNI) (Prost and Jeuffroy 2007; Debaeke et al. 2006; Dordas, 2011). NNI can be calculated by dividing the actual N concentration by the critical N concentration (Nc). The critical N concentration is defined as the minimum N concentration in the shoot biomass required for maximum growth rate, has been established for barley Justes et al. (1994; Nc = 5.35 Ã- Wâˆ’0.442 where W is the total shoot biomass expressed in Mg DM haâˆ’1). The NNI is considered a reference tool for assessing plant N status. However, NNI has a major limitation at the farm level there is a need to determine the actual crop biomass and its N concentration at different growth stages which quite difficult in some cases and there is a need to simplify the evaluation of crop N status with a quicker method of estimating NNI is needed. That's why chlorophyll measurements have been proposed as an alternative tool for estimating the crop N status. There are studies which showed that although CM and RCM readings were related to NNI, but they did not provide a valid and robust estimation of the plant N status because the relationships of CM and RCM readings with NNI varied with sites and years (Ziadi et al. 2008). In addition, CM, RCM, and NNI were not used in barley to determine the N status of the crop and also their relationship to NUE.
Always on Time
Marked to Standard
The Î diagnostic tools that were developed primarily to optimise grain yield. However, the problems of the sustainability of agriculture both from the environmental and the economic points of view require a reconsideration of these factors when calculating fertiliser requirements and especially N fertilizers as they are responsible for an important part of agriculture related pollution through leaching or denitrification (Fageria and Baligar 2005). This involves modifying these diagnostic tools and the decision rules, which are decided on the basis of empirical databases. Therefore, limited pollution risks could be achieved either with low fertiliser rates or cultivars that better absorb and utilize N (Marino et al. 2004). Concerning N, high revenue should be obtained with a maximum yield and quality per unit of N applied. Plant breeding programmes must produce varieties that absorb N more efficiently and use it more efficiently to produce grain (Moll et al. 1982; Dhugga and Waines 1989). Field experiments have shown that genetic variability for N uptake exists in small grains (Loffler et al. 1985; Van Sanford and MacKown 1986; Moll et al. 1982; Dhugga and Waines 1989). This information can be used by the growers for adopting the appropriate cultural practices and also by the breeders for choosing the most efficient selection criteria in order to improve N exploitation. However, such information is limited for barley.
The main objectives of this study were: (i) to establish the relationship between CM and RCM readings and NNI for barley and (ii) to compare both methods as diagnostic tools for predicting yield response to N fertilization, (iii) to determine NUE and its components N utilization efficiency and N uptake efficiency using different barley cultivars.
2. Materials and methods
2.1 Experimental site, cropping history and setup
Field experiments were conducted at the experimental farm of the Aristotle University of Thessaloniki in northern Greece (22o59'6.17" E, 40o32'9.32" N) during the 2003-2004 and 2004-2005 growing seasons (referred hereafter as 2004 and 2005 respectively). The different cultivars that were used to compare the different diagnostic tools for barley and NUE were as follows: Carina (a late-flowering cultivar used for malting), Thessaloniki (an early-flowering cultivar), Konstantinos (a late-flowering cultivar), and Mucho (an early-flowering cultivar and the only one that was six-rowed). The choice of the different cycle cultivars was selected to determine how the flowering date and other characteristics can affect chlorophyll meter and relative chlorophyll meter readings, NNI, and NUE. The different cultivars were chosen in order to have cultivars with different flowering date, different utilization and also to have a two row and six row cultivars. The soil type where the experiment took place was a calcareous sandy loam (Typic Xerorthent), and the soil was sampled pre-planting at a depth of 30 cm and before the application of the fertilizers. The soil contained 7.2 g kg-1 organic matter, 60 kg ha-1 of N-NO3, 26 kg ha-1 of P, and 198 kg ha-1 of exchangeable K and had a pH of 7.96 (1:2 water). The soil characteristics were determined according to methods detailed by Sparks et al. (1996). The preceding crop was durum wheat (Triticum turgidum subsp. durum L.). Weather data (i.e., rainfall, maximum, minimum, and average temperatures) were recorded daily in the experimental area (20 m away from the experimental site) and are given in Table 1 (reported as mean monthly data for the two years of the study) together with the thirty-year averages for temperature and rainfall. During 2004, the spring was quite mild and there was more rainfall during the summer. In contrast, 2005 was warm during the spring and there was less rainfall during the spring (Table 1).
2.2 Crop management and experimental design
The experimental design was split-split plot with the cultivars as the main plots, the fertilizer treatments as the split plot and the source-sink treatments as the split-split plots with five replications. The experimental plots were 3 by 5 meters. The treatments were as follows: 0, 60, and 120 kg N ha-1 were applied (pre-planting) in the form of (NH4)2SO4 (N-P-K, 20.5-0-0). In addition, P and K were applied at a rate of 60 kg ha-1 and 100 kg ha-1 (pre-planting) in the form of superphosphate and K2SO4, respectively and were incorporated in the soil before sowing.
The fertilizer was incorporated with a tandem harrow disc to a depth of 12-15 cm after application. Barley cultivars were sown on 10th of December 2003 and 5th of December 2004 at a rate of 200 kg ha-1 which corresponds to a rate of 370 seeds m-2 with a commercial seeder. Plants were grown without supplemental irrigation in both the growing seasons. The crop was kept free of weeds by hand hoeing when necessary.
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
2.3 Dry matter, nitrogen and grain quality characteristics
The following variables were determined: total aboveground biomass and chlorophyll content at anthesis (Zadoks growth stage 65), and at harvest (Zadoks growth stage 95) (Zadoks et al. 1974). At each sampling, 2 m rows were randomly selected and dried at 80oC until they reached a constant weight. The dry vegetative samples were first ground in a hammer mill and then reground finely using a 1 mm screen. N content was determined by the Kjeldhal method (Dordas and Sioulas 2009).
One thousand grain weight (TGW) was determined by measuring the weight of 100 seeds from each plot and multiplying by 10 in order to express one thousand seed weight. Grain weight per unit volume (kg m-3) was determined by measuring the weight of 0.5 l of grains from each plot. Seed protein content was determined by multiplying by 6.25 the seed N concentration and protein yield was determined by multiplying the seed yield by the seed protein content. Seed yield was determined by harvesting the five central rows with a research plot combine (Wintersteiger AG, Austria) in the last week of June in both years. Relative yields at each plot were computed as the ratio of seed yield at a given N rate by the highest seed yield among all N treatments.
2.4 Chlorophyll measurements
Chlorophyll meter readings were taken with a hand-held dual-wavelength meter (SPAD 502, Chlorophyll meter, Minolta Camera Co., Ltd., Japan). For each plot the 20 youngest fully expanded leaves per plot were used when the plants were at anthesis (Zadoks growth stage 65), at milk stage (Zadoks growth stage 75) and at soft dough stage (Zadoks growth stage 85) (Zadoks et al. 1974). The instrument stored and automatically averaged these readings to generate one reading per plot. Relative chlorophyll meter (RCM) readings were calculated by dividing any SPAD reading by the maximal value from the 120 kg N ha-1. This index, ranging from 0.5 to 1, is also called the sufficiency index (Varvel et al. 1997).
2.5 Nitrogen Nutrition Index
The Nitrogen Nutrition Index (NNI) of the crop at each sampling date was determined by dividing the N concentration of the shoot biomass by the critical N concentration (Nc) (Ziadi et al. 2008). Critical N concentration, the minimum N concentration required to achieve maximum shoot growth, was defined as a function of shoot biomass as proposed for barley by Justes et al. (1994; Nc = 5.35 Ã- Wâˆ’0.442 where W is the total shoot biomass expressed in Mg DM haâˆ’1).
2.6 Nitrogen efficiency and its components
Nitrogen use efficiency was defined as grain production per unit of N available in the soil. Nitrogen use efficiency is Gw/Ns (kg kg-1) in which Gw is the grain weight and Ns is N supply expressed in the same units (e.g. kg ha-1). Ns was calculated from the soil N concentration and N applied rate. There are two primary components of N use efficiency (1) N uptake efficiency (Nt/Ns) (kg kg-1) and (2) N utilization efficiency (kg kg-1) which describes how the N that is absorbed is utilized to produce grains (Gw/Nt), where Nt is the total N in the plant at maturity. Therefore the N use efficiency can follow the equation:
The expression can be expanded to include additional factors. For example, N uptake during grain filling and translocation of N to grain.
where Gw/Ng=grain produced per unit of grain N
Ng/Nt=fraction of total N that is translocated to grain
Na/Nt=fraction of total N that is accumulated after anthesis
Ng/Na=ratio of N translocated to grain to N accumulated after anthesis.
where Ng is the N uptake by grains and Na is the N uptake after anthesis.
2.7 Component analysis
Various expressions were constructed and analyzed according to the method suggested by Moll et al. (1982) and Dhugga and Waines (1989). The analysis involves linearizing the multiplicative relationships by taking logs and then determining the contribution of each component trait to the sum of squares of the resultant trait. The sum of cross products of each component trait by the resultant trait (ï“xiyi) divided by the sum of squares of the resultant trait (ï“yi2) gives the relative contribution of each component variable to resultant variable. This analysis describes the net contribution of each component variable both directly and indirectly through the other variable (Moll et al. 1982). The following expressions were analyzed:
log(N use efficiency (Grainw/Ns))= log(Uptake efficiency (Nt/Ns)) + log(Utilization efficiency (Grainw/Nt))
log(protein yield) = log(grain yield) + log(grain N concentration)
The data were analyzed by the ANOVA method according to a 2Ã-3Ã-4 factorial design (Growing season Ã- N levels Ã- Cultivars) with 5 replications per treatment combination. More specifically, the experiment was set up as a Randomized Complete Block Design for the Cultivars (main plots), and N levels as split plots. A combined analysis over Growing season was carried out according to the aforementioned design Steel et al. (1997). Tukey's post hoc procedure was used for testing the differences between treatment means. The significance level of all hypotheses testing was preset at P<0.05. All statistical analyses were performed using the SPSS ver. 17 software package (SPSS Inc., USA, IL: Chicago).
Cultivars affected most of the characteristics that were studied except from the N concentration at harvest, seed protein content and fraction of total N that is accumulated after anthesis (Table 2). Also N levels affected most of the characteristics that were studied except from the grain volume, the fraction of total N that is translocated to grain, the fraction of total N that is accumulated after anthesis, and the ratio of N translocated to grain N accumulated after anthesis. Growing season affected most of the characteristics that were studied except from the TGW, CM readings at anthesis and at milk growth stage and the respective RCM readings and the fraction of total N that is accumulated after anthesis. The interaction between the cultivars and treatments were significant in the grain volume, the CM readings at anthesis, milk growth stage, and soft dough growth stage and also at the N use efficiency, N uptake efficiency and ratio of N translocated to grain to N accumulated after anthesis. The interaction between the growing season and treatments were significant in protein yield, grain and relative grain yield, the CM and RCM readings at anthesis, at milk stage and at soft dough stage, the N use efficiency and its components. The interaction between the cultivars and growing season were significant in most characteristics and except from N concentration at harvest, the grain protein content, and grain produced per unit of grain N. The interaction among cultivars, treatments, and growing season were significant at the CM and RCM readings at anthesis, at milk stage and at soft dough stage and also grain volume (Table 2). Therefore, these characteristics are presented in greater detail, whereas the rest of the characteristics where there is no interaction among the treatments, years, and cultivars only the main effects are presented.
3.1 Nitrogen concentration and NNI
Nitrogen concentration at anthesis was affected by the cultivar at anthesis but it was not affected by the cultivar at harvest (Table 3). N fertilization affected N concentration at both growth stages and was higher by an average of 8% at both growth stages. Also N concentration was higher during 2005 due to lower dry matter and growth. N concentration in barley plants was not different between the applied N rates but it was different between the control and 120 kg N ha-1.
Nitrogen nutrition indices varied from 0.75 to 1.03 across growing season, growth stage, treatments, and cultivar and was affected by the fertilization level (Table 3). Values of NNI â‰¥ 1.0 indicate that N supply to the crop is nonlimiting or in excess, while values of NNI < 1.0 indicate N deficiency. NNI was higher at Thessaloniki and lower at Konstantinos at harvest and at anthesis it was higher at Much and lower at Konstantinos and Carina. NNI was generally significantly affected by N fertilization as at anthesis NNI were higher by an average of 23% at the fertilization treatments compared with the control and also NNI was higher during the 2005 than the 2004. However, this trend was changed at harvest as NNI was higher during 2004 compared with 2005.
3.2 Grain yield, relative grain yield, and grain quality characteristics
Grain yield was higher at Thessaloniki and Mucho and followed by Carina and Konstantinos. Also grain yield was increased with N fertilization in the three cultivars by an average of 27% (Table 4). Also grain yield was higher during the first year due to the better weather conditions. Relative grain yield was lower at Carina and Konstantinos cultivars and higher at Thessaloniki cultivar and ranged between 0.74 to 1.0. Relative grain yield was increased by 26% with N fertilization compared with the control. Also at 2004 the relative grain yield was 0.90 and was lower during the second year which was 0.84.
Grain protein content was not affected by the cultivar but was affected by the N level and growing season. In particular, grain protein content was higher by an average of 6% at the N levels compared with the control and there was no difference between the two N rates. Also it was higher during the 2004 growing season compared with the 2005. A slightly different trend was found at the protein yield as it was lower in Konstantinos and Carina due possible to the lower grain yield and higher at the other two cultivars (Thessaloniki and Mucho) (Table 4). Also there was an increase at the protein yield with N application and was higher during 2004 compared with the 2005 growing season. Grain volume was higher at Thessaloniki and Mucho two cultivars that were early flowering and was followed by Konstantinos and the last one was Carina (Figure 1). However, the N level did not affect the grain volume and also the growing season (Table 2). TGW was lower at Carina and Konstantinos cultivars and higher at Thessaloniki and Mucho (Table 4). Also TGW was higher at the N fertilization level and was not affected by the growing season (Table 4).
3.3 Chlorophyll and relative chlorophyll meter readings
Chlorophyll meter (CM) readings were affected by the N treatments and were higher at both N levels compared with the control (Figure 2). The same trend was observed during the second year. At anthesis, CM readings were higher by an average of 15 % in the fertilized treatments compared with the control. At milk growth stage, there were significant differences between the control and the N treatments especially during the first year, where the chlorophyll level was by 22% higher compared with the control. RCM readings were higher at the fertilization treatments compared with the control and were in the range of 0.76 to 1.00. At anthesis, RCM readings were higher by an average of 24 % in the fertilized treatments compared with the control. At soft dough stage, there were significant differences between the control and the N treatments especially during the first year, where the RCM level was by 22% higher compared with the control (Figure 3).
3.4 Nitrogen efficiency and its components
Nitrogen use efficiency and its components N uptake efficiency and N utilisation efficiency were affected by N fertilization, growing season, and cultivars. N use efficiency, N uptake efficiency, and N utilization efficiency was higher at the control compared with the two fertilization treatments (Table 5). Thessaloniki had higher N uptake efficiency at both N levels, while Konstantinos had the lowest and it was in the range from 0.67 to 1.14. Konstantinos had highest N utilization efficiency compared with Thessaloniki and Carina. N utilisation efficiency ranged from 26.87 to 41.14 kg kg-1 N. The fraction of total N that was translocated to grain (Ng/Nt) was affected only by the cultivar and growing season and was higher in Thessaloniki compared with the other cultivars and also it was higher during 2005 compared with the 2004 (Table 5). The fraction of total N that was accumulated after anthesis was higher at Carina (a late flowering cultivar) compared with the Thessaloniki cultivar (an early flowering cultivar). Also it was higher at the control compared with the 120 kg N ha-1 and was higher during 2005 than in 2004. The ratio of N translocated to grain to N accumulated after anthesis (Ng/Na) was higher at the Thessaloniki and lower at the Carina cultivar and it was higher during the 2005 growing season compared with the 2004. Grains produced per unit of grain N were higher at the Mucho and Carina compared with the other two cultivars. In addition, grains produced per unit of grain N were higher at the control compared with the other two treatments and also during the 2004 compared with the 2005.
3.5 Component analysis of the different traits.
The relative contributions of N use efficiency components are presented in Table 6. N utilization efficiency accounted for more of the variation of N use efficiency than the N uptake efficiency and was higher at the 0 kg N ha-1 than two N levels and especially during the first year. During the 2005 the trend was different as the N uptake efficiency accounted for more of the variation than the N utilization efficiency and also there was at 120 kg ha-1 higher variation in the N utilization efficiency than the N uptake efficiency. The variation attributed to the N uptake efficiency was higher at the control which indicates that there was a significant interaction in N use efficiency among cultivars, treatments, and years.
The relative contributions of grain protein yield components showed that protein concentration was accounted more of the variation in grain protein concentration that grain yield. The variation attributed to the protein concentration was more at the control than at the other N levels especially during the 2004 growing season and there was no difference between the N levels (Table 6).
Barley is a species that has been studied extensively however, there are many characteristics such as CM and RCM readings, NNI and NUE that are used to describe responses to different treatments and their relationships are not known (Table 7). CM readings at anthesis were correlated with N concentration at anthesis and negatively correlated with NUE and N uptake efficiency (Table 7). Also N concentration at anthesis was correlated with NNI at anthesis, TGW, and negative correlated with NUE, and N uptake efficiency. Grain yield was correlated with NNI at harvest, grain protein content, TGW, NUE, N uptake, and utilization efficiency. NUE was correlated with grain protein content, N uptake efficiency and N utilization efficiency. Grain protein content was correlated with NNI at harvest and grain yield. NNI was correlated with the N concentration at the same growth stage also NNI at harvest was correlated with grain yield. NNI was negatively correlated with NUE and N utilization efficiency. This is the first report where the effect of N supply on certain physiological characteristics was determined for barley and also their relationship with grain yield.
4.1 Nitrogen concentration and NNI
Nitrogen fertilization affected N concentration of barley plants at anthesis and at harvest. When the N level is marginal, as in the present study, there was an increase in N concentration with N fertilization (Dordas et al. 2008). However, even if the soil N concentration is marginal there are studies which showed no increase in N concentration and also in grain yield (Prystupa et al. 2004; Le Gouis et al. 1999) which can be because the response of barley to N fertilization is affected by the soil type, barley cultivar, climate, growing season moisture conditions, N fertilizer form and placement, and seeding rate and seeding date (Prystupa et al. 2004; Le Gouis et al. 1999).
The main objective of this study was to determine whether there is a relationship between NNI and N concentration and also CM readings. NNI varied from 0.75 to 1.03 across growing season, cultivars, N treatments, and developmental stage. Similar ranges have been reported for a number of different crops such corn (Ziadi et al. 2008; 0.29-1.3; Plénet and Cruz 1997; 0.55-1.45; and Justes et al. 1997; 0.45-1.30), annual ryegrass (Marino et al. 2004; 0.4-1.6), spring wheat (Ziadi et al. 2010; 0.34-1.43), durum wheat (Debaeue et al. 2006; 0.25-1.5), cotton (Xiaping et al. 2007; 0.75-1.16), and linseed (Dordas 2011; 0.65-1.16). Values of NNI â‰¥ 1.0 indicate that N supply to the crop is nonlimiting or in excess, while values of NNI < 1.0 indicate N deficiency. In agreements with previous findings in other crop species NNI for barley showed relatively low variation within N treatments through the growth period and showed higher values at higher N fertilization level (Lemaire et al. 2008). The increase in NNI values with increasing N fertilization has been reported in corn, wheat, linseed, and other crops but has not reported in barley (Plénet and Cruz 1997; Justes et al. 1997; Dordas 2011). NNI is recognized as a reference method for detecting N deficiency in wheat in Europe and from the present study seems that can be used also for barley (Justes et al. 1997). NNI can be used as a priori diagnosis of plant status during crop growth to determine the necessity of applying additional fertilization. However, a major difficulty in using the NNI as a diagnostic tool is the need to determine the actual crop biomass and its N concentration (Lemaire and Gastal 2009). Therefore, it was suggested for many crops and seems that the same exists for barley that NNI can be used as a reference for simpler procedures (eg chlorophyll measurements or nitrate concentration in stem base extract) to determine crop N status (Justes et al. 1997; Debaeke et al. 2006; Prost and Jeuffroy 2007; Dordas 2011). The NNI was also used in crop models to account for the effect of N on growth and yield of winter wheat (Devienne-Barter et al. 2000).
4.2 Grain yield, relative grain yield, protein content and protein yield
Grain yield was increased by an average of 27 %, in agreement with other studies in which the N was limited and the N application was shown to increase grain yield (Dordas and Sioulas 2008; Prystupa et al. 2004; Dordas 2011; Papakosta and Ganianas 1991; Delogu et al. 1998; Maidl et al. 1998). Relative grain yield was also increased by an average of 26%, however the relative grain yield was not studied before in barley and it was not used to determine the response to N fertilization. The response of grain yield to N fertilization can also vary with site, year, soil type, cultivar, climate, N form and placement, seeding rate, and date (Papakosta and Ganianas 1991; Maidl et al. 1998 ). During 2004, the higher grain yield in the present study was due to the better weather conditions. Low rainfall and high temperatures during anthesis and grain filling stage can have a significant effect on grain yield (Bloom et al. 1985). In addition, observed limited response of barley to N application where soil NO3 levels were high. However, when the N is applied in high rates grain yield can be reduced because of lodging (Maidl et al. 1998; White 1995).
Grain protein content was increased with N application by an average of 6 % in all cultivars over the two years and grain protein yield was increased by an average of 28% compared with the control. The highest increase in grain protein yield compared with the grain protein content was because N fertilization affected more the grain yield than the grain protein content. Grain protein content was found to increase with N fertilization (Maidl et al. 1998; Gauer et al. 1992; Ehdaie and Waines 2001) when the weather conditions are favorable. But when there was low rainfall and with high temperatures there was no response to N fertilization ( ). However, the TGW was affected by N fertilization. Several other authors have also reported that N fertilization can affect TGW ( ). However, others found that N fertilization did not increase the TGW (Arduini et al. 2006; Ferrise et al. 2010; Gonzales et al. 2003). Grain volume was affected by the N treatments and showed a significant interaction among the growing season x treatments x cultivars.
4.3 Chlorophyll meter and relative chlorophyll meter readings
Chlorophyll meter readings ranged from 29 to 55 which is quite typical since there is variability in chlorophyll content (Ziadi et al. 2008; Dordas et al. 2008). Chlorophyll meter readings were affected by the N treatment and were higher at anthesis and milk stage than at dough stage. CM readings generally show increase during the growing season up to a maximum and then gradually decrease when the leaf senescence start and there is degradation of the chlorophyll content (Ziadi et al. 2008; Dordas et al. 2008). However, this variation over the growing period is important as there is a need to specify the developmental stage at which CM readings are taken. At the earlier growing stages the CM readings are not significantly affected by N treatments which can be because of the residual N in the soil and the low requirements of N for the plant (Ziadi et al. 2008; Dordas et al. 2008). As the plant grows and especially at anthesis is generally a time where the differences in the response to N fertilization are more pronounced. Furthermore as the plants reach maturity there is loss of chlorophyll due to senescence and there is a gradual decrease of the CM readings. There was much higher decline in CM readings at the control than at the N treatments which indicates that the lower amount of N available to the control was perhaps remobilized for grain growth, causing the leaves to senesce quicker and lowering the amount of chlorophyll (Shukla et al. 2004). This clearly indicates that when there is adequate N supply in the soil leaf senescence is slower and the plant supplies the grain with N and photoassimilates for longer time which results in higher yields (Eghball and Power 1999). The CM readings can be affected by cultivar, site characteristics, developmental stage, disease or insect damage, plant density and other nutrient deficiencies (Masoni et al. 1996). That's why RCM readings have been proposed which are recommended to account for the influence of the above mentioned factors on CM readings (Blackmer and Schepers 1995; Piekielek et al. 1995; Waskom et al. 1996). RCM readings ranged from 0.76 to 1.00 and the variations that was noted was lower at the RCM readings compared with the CM readings which agrees with others (Ziadi et al. 2008).
4.4 Nitrogen efficiency and its components
Nitrogen use efficiency and its components were generally higher in 2004 than in 2005. NUE was decreased with increasing N fertilization rate (Marino et al. 2004). Genotype differences were observed in NUEs with Thessaloniki to have the highest value and the lowest was found at Carina. NUE was not correlated with seed yield but it was negatively correlated with the relative grain yield, indicating that high yield was not associated with more efficient exploitation of N. These relationships suggest that high yield is the result of better exploitation of N or high seed N concentration, and consequently, may be accompanied by low NUEs (Le Gouis et al. 1999; Sinebo et al. 2004). Therefore, barley breeders should select for both high yield and NUEs in order to ensure an improvement in both traits (Le Gouis et al. 1999).
Nitrogen uptake efficiency was higher for 2005 than for 2004. Under field conditions, soil N availability shows high spatial and temporal heterogeneity that affects plants' N uptake. During the winter and early spring of 2005 lower rainfall during grain filling period and relatively high mean temperature (especially in April and early May, which is anthesis and the grain filling starts) could have determined short periods of soil water deficiency and, consequently, a decreased in N uptake of the applied fertilizer (Bloom et al. 1985). In contrast, the higher water availability during the experimental period of 2004 in relation to the same period of 2005 might have favored a greater N uptake from the fertilizer applied. The uptake rate of a given element depends on its external concentration and on the plants absorption capacity (Lee 1993). It was found that the maximal nutrient absorption capacity of plants is higher than that required to obtain the maximum yield (Jarvis and Macduff 1989; Jeuffroy and Meynard 1997). Justes et al. (1994) observed that, for a certain amounts of aerial biomass, the N concentration could be up to 160% of the N concentration considered critical. Nutrients such as N can be accumulated (stored) in plants during periods of external abundance and consumed in subsequent growth when they are externally limited (Bloom et al. 1985).
Higher values of N utilization efficiency were found in 2004 than in 2005. This means that a higher amount of biomass and grain yield per unit of N uptake was produced in 2004. According to the previous discussion, the lower N utilization efficiency during 2005 in relation to 2004 reflects a luxury consumption. In other words, in 2004, plants acquired N in excess for its current growth.
4.6 Component analysis of the different traits
Analysis of the different components (N uptake efficiency and N utilization efficiency) that affect NUE showed significant differences in the magnitude of the contribution of each component to the variation in NUE among genotypes and also among the N treatments. Nitrogen utilization efficiency was the most important components of NUE and accounted for more of the variation of N use efficiency than the N uptake efficiency and was higher at the 0 kg N ha-1 than the 120 kg N ha-1 in both years. Like Moll et al. (1982), Dhugga and Waines (1989), Ortiz-Monasterio et al. (1997), and the present study showed that the contribution of N uptake efficiency and seed N utilisation efficiency were dependent on N level. Ortiz-Monasterio et al. (1997), found that N uptake efficiency accounted more for the variation in N use efficiency at the control than at the N fertilization treatments which disagrees with the present study where the N utilization efficiency accounted more for the variation of the N use efficiency. When N is rare, the ability to absorb N is certainly of paramount importance and would then be related to root characteristics. It may be hypothesised that differences for the ability to explore the soil or to absorb N existed in the material that were tested. When N is not the limiting factor, N utilisation efficiency have to be more determinant as N will be available for each genotype independent of the efficiency of their root system. Ortiz-Monasterio et al. (1997) proposed selecting in medium-high fertility environments to improve for both low and high fertility conditions. NUE was negatively correlated with N concentration at anthesis, suggesting that low N concentration may be indication of higher NUE.
Barley is a species that has been studied extensively but there are many characteristics that relate to NUE and NNI that were not determined and are important for higher productivity of the barley cultivars and also for the selection of new cultivars. Therefore, it is important to know whether the selected characteristics that it was chosen in this study can be used to describe the response of other cultivars to N deficiency, or whether we can use them to improve selection of new cultivars and hence increase productivity. It was found a linear relationship between CM readings at anthesis, N concentration at anthesis, and negative correlation with NUE and N uptake efficiency. NUE was correlated with grain yield, indicating a need for more research for a better understanding of those characteristics and a need to try to increase both characteristics in new cultivars also this indicated that high yield was associated with more efficient exploitation of N. Positive correlation was found in other studies (Sinebo et al. 2004). However, other reported an negative correlation between NUE and seed yield in linseed (Dordas 2011). Additional correlation analysis indicated a negative correlation between NUE and NNI at anthesis and at harvest. These relationships suggest that high yield is the result of better exploitation of N and as NNI increases this is accompanied by lower NUEs. In this study the effect of N supply on certain physiological characteristics was determined for barley and also their relationship with grain yield. It is obvious that CM can be used as a tool for selection of new cultivars with high yield. There are many trade-offs between the different components of the NUE and many factors that can affect them, so correlation analysis shows the general trends of these characteristics. However, more research is needed to explore these tools for barley breeding and also for better barley management especially under rainfed conditions.
Nitrogen is one of the most important nutrients needed for plant growth and development. Nitrogen fertilization affected N concentration of the barley plants at anthesis and at harvest and also grain protein content and grain protein yield. CM readings were affected by the N treatment and it was higher at both N levels compared with the control. NNI varied from 0.75 to 1.03 across years, growth stage and cultivar and was affected by the N level. In addition, N use efficiency, N uptake efficiency and N utilization efficiency was higher at the control compared with the two fertilization treatments. N utilization efficiency accounted for more of the variation of N use efficiency than the N uptake efficiency and was higher at the 0 kg N ha-1 than the 120 kg N ha-1 in both years. CM readings at anthesis were correlated with NNI at anthesis, and grain yield. NUE and its components were negatively correlated with CM readings. In conclusion, the interrelations found among the various NUE-related traits suggest that using simple selection criteria to improve NUE of barley might have negative implications on grain yield and quality. Therefore, evaluation and selection of different genotypes for NUE should be based on multiple criteria rather than just one criterion and also should be accompanied by evaluation for grain yield.
The author is grateful to Professors N. Fotiadis and A. Gagianas, Faculty of Agriculture, Aristotle University of Thessaloniki, for their critical review of the manuscript.