Timing and growing conditions of nitrogen topdressing influence the grain yield and protein content of four wheat cultivars

Bragantia, Campinas, v. 78, n. 3, p.361-370, 2019 ABSTRACT: The application of an adequate rate and splitting of nitrogen is essential for wheat grain yield and protein content. The aim of this work was to adjust nitrogen management approaches regarding agronomic performance and protein content of wheat cultivars in various environments. Field experiments were conducted under no-tillage system on soybean mulch during the 2011 and 2012 growing seasons in Londrina and Pato Branco regions. The experimental design was a randomized block in split plot with four replicates. Four wheat cultivars (IPR Catuara TM, BRS Gaivota, Quartzo, CD 120) were tested with six nitrogen (N) management forms. Were evaluated: number of ears per unit area (NEA); plant height (PH); thousand-kernel weight (TKW); test weight (TW); grain yield (GY); and protein content (PC). The combined ANOVA (p ≤ 0.01) and CROP PRODUCTION AND MANAGEMENT Article


INTRODUCTION
Wheat (Triticum aestivum L.) is one of the world's most important cereals with over 700 million tons produced each year. In Brazil, approximately 5.4 million tons, corresponding to 50% of Brazilian consumption of this cereal. Paraná state contributes to 50% the amount of the wheat produced in this country, with an average production of 2700 kg.ha -1 . Moreover, there are differences in consumer preferences, implying the need for wheat production with suitable quality characteristics for baking (Brum and Müller 2008).
Various factors limit wheat grain yield and grain quality, including the management of nitrogen (N), the environment and the cultivar. N can influence wheat grain yield (Benin et al. 2012a), yield components (Trindade et al. 2006;Benin et al. 2012a) and wheat grain quality (Pinnow et al. 2013). Similarly, the environment has a significant effect on the grain yield (Silva et al. 2011) and aspects of grain quality such as protein content (Silva et al. 2014). The protein content of wheat grain is an important parameter in the analysis of the quality attributes of wheat. Any fluctuation in protein content significantly affects its technological quality (Hrušková and Faměra 2003).
However, the dynamism of N and its tendency to losses creates a challenging environment to manage this nutrient in topdressing efficiently (Fageria and Baligar 2005), mainly due to various reactions and instability in the soil. The low efficiency of N is attributed to ammonia volatilization (Van der Gon and Bleeker 2005), leaching, surface runoff and denitrification (Fageria and Baligar 2005). Therefore, improving the efficiency of N use by plants is crucial to enhance grain yield, reduce production costs and improve environmental quality (Fageria and Baligar 2005).
Amidic N sources (urea) have higher loss by volatilization when applied to soils of low cation exchange capacity (CEC), basic pH, low humidity and high temperature. The flow of ammonia volatilization begins immediately after application of urea by rapid hydrolysis in soil. Moreover, ammonium sulfate and ammonium nitrate do not result in volatilization losses, even when surface-applied. Further, the energy requirement for ammonium assimilation is lower than for nitrate assimilation, because the former does not need to be reduced for incorporation into amino acids (Magalhães et al. 1993). Nevertheless, urea is one of the most frequently used N sources in Brazil, based on its high N concentration and low cost per unit of nitrogen.
An important practice in crop management is to supply the plant during critical periods by using splitting N fertilizers and applying a suitable amount of N. The correct timing of application mitigates the loss of nutrients and increases the efficiency of nutrient, improving grain yield, yield components (Basso et al. 2010) and grain quality parameters (Fuertes-Mendizábal et al. 2010).
There are conflicting reports on the ideal time to N application or splitting of nitrogen in the wheat crop (Teixeira Filho et al. 2010). In Brazil the late application of N fertilizer to wheat is often criticized because the accumulated protein might not be functional, which implies lower gains in gluten strength. The industrial quality of wheat genotypes for Brazilians is not usually measured based on the protein content in grain (Vázquez et al. 2012).
The efficiency of N also varies depending on the cultivars released year by year, which differ in their response to the N applied in topdressing. Moreover, the response to N relies on the growth environment for each cultivar (Benin et al. 2012a). Therefore, it is technically impracticable to generalize the management of N for wheat cultivars (Teixeira Filho et al. 2010).
Consequently, the integrated study of the management of N fertilization of wheat must consider the distinction between environments, N sources, timing of the application, and especially genetic factor. The aim of this study was to adjust nitrogen management approaches with regard to the agronomic performance and protein content of wheat cultivars under different environmental conditions.

MATERIALS AND METHODS
Field experiments were performed during the 2011 and 2012 growing seasons in two wheat-growing regions of Paraná state, Brazil: Londrina (23°22'09" S latitude, 51°10'13" W longitude and 549 m elevation) and Pato Branco (26°09'58" S latitude, 52°42'22" W longitude, altitude 749 m elevation). Hence resulting in four different growth conditions. The predominant climate of both locations is the Cfa type and their soil is classified as Oxisol.
Wheat is grown in three principal regions of Brazil: Central, South Central and South. To evaluate wheat genotypes, these regions are subdivided by water regime, air temperature and elevation above sea level. The categorization of these subdivisions is based on their Value for Cultivation and Use (VCU) categories, as follows: wet, cold and high elevation (VCU 1); wet, moderately warm and low elevation (VCU 2); moderately dry, warm and low elevation (VCU 3); and dry, warm and Cerrado (VCU 4). Londrina is in the VCU 3, and Pato Branco is in the VCU 2.
The experimental design was a randomized block in split plot design with four replicates. Four cultivars of wheat (IPR Catuara TM, BRS Gaivota, Quartzo and CD 120) were tested in six management forms of nitrogen (subplots). The sowing was carried out mechanically under no-tillage system on soybean mulch at both locations. In The management forms of nitrogen were: N1 -without N topdressing; N2 -60 kg.ha -1 of N in the formulation of urea applied at the beginning of tillering; N3 -80 kg.ha -1 of N in the formulation of urea splitted into 60 kg.ha -1 (beginning of the tillering stage) and 20 kg.ha -1 (booting stage); N4 -100 kg.ha -1 of N in the formulation of urea splitted into 60 kg.ha -1 (beginning of the tillering stage) and 40 kg.ha -1 (booting stage); N5 -80 kg.ha -1 of N splitted into 60 kg.ha -1 (beginning of the tillering stage) applied in the formulation of urea, and 20 kg.ha -1 (booting stage) in the formulation of ammonium sulfate; N6 -100 kg.ha -1 of N splitted into 60 kg.ha -1 (beginning of the tillering stage) applied in the formulation of urea, and 40 kg.ha -1 (booting stage) in the formulation of ammonium sulfate.
The N fertilization was performed on specific dates for each cultivar according to the cultivar cycle. In Environment A1, after N fertilization, the experiment was irrigated with 15 mm due to drought at the time of the nitrogen topdressing application. The climatic conditions of each environment are shown in Fig. 1. The cultivars characteristics are shown in Table 1. The chemical soil characteristics of each environment are shown in Table 2.  Oct. Nov.
The following evaluations were performed: a. Plant height (PH) at crop maturity, defined as the distance from soil level to the apex of the spike excluding awns, considering ten plants at random of each subplot; b. Number of ears per unit area (NEA) at crop maturity, defined as the number of ears per linear meter in the row and estimated per area; c. Thousand-kernel weight (TKW), defined by measuring the mass of 250 grains in triplicate (counted electronically) of each subplot; d. Test weight (TW), corresponding to the mass of grains in 100 L; e. Grain yield (GY), determined after mechanical harvesting of the useful area in each subplot, expressed as kg.ha -1 and reported on a 13% moisture content basis; f. Protein content in whole meal (PC), determined by near infrared reflectance (NIR). Data were subjected to joint analysis of variance (p-value ≤ 0.01) and comparison of means by Tukey test (p-value ≤ 0.01).

RESULTS AND DISCUSSION
The low coefficient of variation values indicate that our inferences are reliable with high experimental precision (Table 3). In the same way, in the literature the optimal coefficient of variation differs between the characteristics evaluated (Benin et al. 2012a). Barraclough et al. (2010) and Benin et al. (2012a) have previously reported the interactions observed in analysis of variance ANOVA (Table 3). Mean squares for the environmental effect were the most important, with high magnitude in all the characteristics evaluated, which is typical for experiments involving different environmental conditions (Benin et al. 2012a).
PH was influenced by the interaction between cultivars and environments ( Table 3). The BRS Gaivota and Quartzo cultivars had higher PH in the A1 and A3 environments; whereas that of the IPR Catuara TM and CD 120 cultivars increased in the A1 environment. PH was unaffected by cultivars in environments A1 and A4 (Table 4). In the A2 environment, IPR Catuara TM had the lowest PH, whereas the A3 environment propitiated the smallest PH for the IPR Catuara TM and CD 120 cultivars. All cultivars had the lowest PH in the A4 environment, likely due to water stress. Moreover, in the A1 environment all cultivars showed lodging at the stage of earing, except for BRS Gaivota.
N stimulates vegetative growth and stem elongation. Accordingly Espindula et al. (2010) showed that application of N increases wheat PH. Nevertheless, this study and previous studies found that wheat PH was unaffected by N fertilization (Teixeira Filho et al. 2010). The absence of an effect of N on wheat PH may be due to the N splitted fertilization, mainly because the second N application was performed after the beginning of stem elongation. The occurrence of high temperatures and irregular rainfall in the A4 environment may have contributed to the reduction in PH. Oliveira et al. (2011) reported that high cultivation temperatures negatively affected all plant traits, including PH. NEA was influenced by the cultivars × environments and N × environments interactions (Table 3). All cultivars showed the greatest NEA in the A1 and A3 environments (Table 4). Environments A1 and A4 propitiated the lowest NEA for IPR Catuara TM cultivar, which differed from those of other cultivars. Only in the A3 environment the NEA was affected by the form of N; moreover, the effect of environment differed for each form of N management. The IPR Catuara TM cultivar had the lowest NEA regardless of the cultivation environment and did not differ from BRS Gaivota in environments A2 and A3.
The environmental effect on NEA demonstrates the importance of the suitability of cultivars and nitrogen level to each environment. In the A3 environment, the NEA increased because the fertilizer of N was split at tillering and booting stage (N3, N4 and N6). However, in the A3 environment no significant difference was observed between the sources of N topdressing. Changes in weather, especially under solar radiation, promote changes in stem elongation rate and the internodes of the plant, affecting the development of tillers that will produce ears. When the plant has high N concentration, the critical red:far red (lower production of tillers) is smaller than at lower concentration. The application of N at the earing stage may contribute to the increase in leaf area index, which results in greater photosynthetic area (Sparkes et al. 2006). The greater photosynthetic area contributes in increasing the availability of carbohydrates to maintain production of tillers and ears that result in grains (Almeida et al. 2004), which explains the effect of N management in the A3 environment. This supports the necessity of N application in the development stages recommended for wheat crop, which enhances the maximum exploitation of the genetic potential of cultivars.
Although the NEA is an important component of GY there are doubts regarding its use, because it has high complexity and is influenced by genetics (Benin et al. 2012b), environmental conditions (Benin et al. 2012a), the availability of N (Sparkes et al. 2006;Benin et al. 2012b) and seeding rate (Sparkes et al. 2006). However, the results showed that large NEA increases GY, especially in the dry A4 environment (Table 5), as shown by the cultivar BRS Gaivota. On the other hand, the GY of cultivars with low NEA was negatively influenced under unfavorable environmental conditions (environment A4), as demonstrated by the cultivar IPR Catuara TM.
The TKW was affected by the interaction between environment and cultivars, between N and environments ( Table 3). The A3 environment led to a lower TKW for all cultivars except BRS Gaivota, which was not influenced by environment effect (Table 4). IPR Catuara , N2 (60 kg.ha -1 of N using urea at beginning tillering stage), N3 (60 kg.ha -1 of N at beginning tillering stage and 20 kg.ha -1 of N at booting stage, both using urea), N4 (60 kg.ha -1 of N at beginning tillering stage and 40 kg.ha -1 of N at booting stage, both using urea), N5 (60 kg.ha -1 of N using urea at beginning tillering stage and 20 kg.ha -1 of N using ammonium sulfate at booting stage) and N6 (60 kg.ha -1 of N using urea at beginning tillering stage and 40 kg.ha -1 of N using ammonium sulfate at booting stage).
TM cultivar had the highest TKW in the A1, A2 and A4 environments. Only management without N topdressing was not influenced by environmental effect. There were no differences in TKW with the increment of N topdressing in the environments A1, A2 and A4. The TKW was reduced with N fertilization topdressing in A3 environment. The largest TKW for IPR Catuara TM is related to its low tillering capacity and thus lower compensatory effect of the number of ears per unit area, which increases the TKW (Motzo et al. 2004). The lack of increase in TKW might be due to the plants using the available N to promote tillering and leaf growth.
The interaction between cultivars and environments (Table 3) influenced the TW. The A3 environment induced the lowest TW for all cultivars, but did not differ from environment A4 with regard to BRS Gaivota. The IPR Catuara TM cultivar had the highest TW in the A1, A2 and A4 environments. In the A3 environment no meaningful difference between cultivars for TW was observed. Guarienti et al. (2005) reported that excessive water in the 20 days pre-harvest causes reduction in the TKW; therefore the low TKW in A3 environment. There were successive changes in grain moisture due to rain following a dry period (Fig. 1). Such weather stresses lead to a high respiration rate and an increased consumption of carbohydrate and reserves in the grains (Guarienti et al. 2005).
The cultivar environment interaction and the N environment interaction influenced the GY (Table 3). Only in the A3 environment no meaningful difference was observed in GY between cultivars. The GY of BRS Gaivota and Quartzo cultivars experienced no reduction even when grown in the A4 environment (Table 5). The largest GY for IPR Catuara TM cultivar occurred in the A1 environment. The BRS Gaivota presented adaptability to A1 and A4 environments. There was no significant difference in GY for the CD 120 cultivar in the A1, A2 and A3 environments. The effect of environment influenced the GY regardless of the N management form, and the A1 environment provided the highest GY with all forms of N except at N4. Significant effects were observed in relation to the form of N management only in the A3 environment, in which N4 increased the GY.
Therefore, the cultivars tested had similar responses to N. In three of the four tested environmental conditions, the N management did not change the GY. These results (Table 5) confirm that N depend on environmental conditions and genotype, which makes N a difficult nutrient to manage (Fageria and Baligar 2005;Van der Gon and Bleeker 2005). Furthermore, Barraclough et al. (2010) reported that cultivars that use N efficiently are ideal. N fertilizers are expensive and the applied N might be under utilized in poor growing conditions.
Regardless of the choice of cultivar, the environment influenced the PC. Indeed, the choice of cultivars did not influence the PC in Pato Branco location (A3 and A4 environments) ( Table 6). The environment influenced the N management. However N management did not influence the PC in Pato Branco (A3 and A4). The PC in wheat flour mainly gluten is one of the most important parameters in the analysis of quality attributes of wheat (Pirozi et al. 2008). Despite the positive effects of N on the PC, it was observed that the N did not influence the tested cultivars; the result obtained may be because the ability to accumulate proteins in response to N fertilizer depends on the wheat cultivar (Ercoli et al. 2011). , N2 (60 kg·ha -1 of N using urea at beginning tillering stage), N3 (60 kg·ha -1 of N at beginning tillering stage and 20 kg·ha -1 of N at booting stage, both using urea), N4 (60 kg·ha -1 of N at beginning tillering stage and 40 kg·ha -1 of N at booting stage, both using urea), N5 (60 kg·ha -1 of N using urea at beginning tillering stage and 20 kg·ha -1 of N using ammonium sulfate at booting stage) and N6 (60 kg·ha -1 of N using urea at beginning tillering stage and 40 kg·ha -1 of N using ammonium sulfate at booting stage).
R. R. Silva et al. The variability in PC due to the environment has been previously described by other researchers and may be explained by temperature and precipitation changes (Motzo et al. 2004). Moreover, the environmental effect on the variability in the PC is significantly higher than the genotypic effect (Vázquez et al. 2012). In fact, in the A4 environment has the highest temperature and water stress during grain filling increased the PC. This relationship between climate variables and PC is very debated and uncertain. However, some results indicate that PC increases under low rainfall conditions (Garrido-Lestache et al. 2004).
The regulation of protein content is complex and the effect of rainfall and temperature are not consistent across sites or years. Despite the increased protein in drought conditions, we emphasize that a rise in protein content in grains does not always result in higher-quality bread, because the synthesis of gliadins usually increases (Hajheidari et al. 2007). These protein subunits do not contribute significantly to gluten strength. Although the previous crop in the environments in this study was soybean, the analysis of PC also depends on the C/N ratio of mulch straw (Pinnow et al. 2013). Thus, under conditions that limit the availability of N, protein synthesis declines and grain starch increases.
PC was negatively correlated (r = -0.90; p ≤ 0.05) with wheat GY. Previous studies have reported negative relationship between PC and GY (Šíp et al. 2013) and concomitant increase in PC and GY (Pinnow et al. 2013). Therefore, the correlation between PT and GY depends on factors as environment, genotype and management crop (Pinnow et al. 2013;Šíp et al. 2013). It is impossible to make generalizations about this relationship.
Thus, in environments A1, A2 and A4 there are evidences that it is not necessary to use nitrogen topdressing when the cultivars with IPR Catuara TM are used in the environment A1, Quartzo cultivar in A2, environment and BRS Gaivota cultivar in A4. Moreover, in the A3 environment, this study recommends using 100 kg.ha -1 of splitted N (N4) in the topdressing using urea, regardless of the cultivar. However, due to difficulties related to the recommendation of N fertilization and climate prediction, it is often preferable to use cultivars with high GY without N in topdressing (Barraclough et al. 2010). Another important result is that in three of the four tested growth environments, a nonsignificant difference in GY between the N sources was observed, perhaps associated with good soybean mulch on the soil. Furthermore, when different N sources (booting stage) were applied, the crop lines were already closed reducing the temperature and retaining soil moisture (Da Ros et al. 2005).

CONCLUSION
The interaction between cultivars and growing conditions influences all yield components grain yield and protein content. The interaction management forms of N and environments influenced the thousand-kernel weight, number of ears per area, grain yield and protein content. In environments where there is low rainfall, N topdressing can be suppressed without influencing grain yield and protein content. In ideal weather conditions the wheat grain yield is enhanced with the application of 60 kg.ha -1 of N at the beginning of tillering and 20 kg.ha -1 of N at booting. The use of appropriate cultivars for each environment is essential in achieving high , N2 (60 kg·ha -1 of N using urea at beginning tillering stage), N3 (60 kg·ha -1 of N at beginning tillering stage and 20 kg·ha -1 of N at booting stage, both using urea), N4 (60 kg.ha -1 of N at beginning tillering stage and 40 kg·ha -1 of N at booting stage, both using urea), N5 (60 kg.ha -1 of N using urea at beginning tillering stage and 20 kg·ha -1 of N using ammonium sulfate at booting stage) and N6 (60 kg·ha -1 of N using urea at beginning tillering stage and 40 kg·ha -1 of N using ammonium sulfate at booting stage).
production level. Th e forms of N applied as topdressing did not infl uence the protein content of cultivars in Pato Branco location.