Effects of row spacing and nitrogen topdressing fertilization on the yield of upland rice in a no-tillage system

This study aimed to evaluate the effect of row spacing and nitrogen topdressing fertilization of two materials (genotype 07SEQCL441 CL and cultivar BRS Esmeralda) on the plant height, yield components, grain yield, and quality of an upland rice crop grown in a no-tillage system. Trials were conducted for two growing seasons under field conditions in a 3 x 4 factorial, randomized, complete block design, with four replications. For each material, treatments consisted of the combination of row spacing (0.225, 0.35, and 0.45 m) with nitrogen (N) applied as topdressing (0, 50, 100, and 150 kg ha). The lowest row spacing (0.225 m) for genotypes 07SEQCL441 CL and BRS Esmeralda provided a higher number of tillers, number of panicles m, and grain yield of rice. Increasing rates of N in the topdressing improved the rice grain yield for both cultivars, but for 07SEQCL441 CL, the grain yield was positively affected only to applications up to 50 kg N ha. Row spacing and N rates did not affect the rice grain quality. Therefore, these results indicate that the narrowest row spacing used (0.225 m) with N fertilization as topdressing increased the rice grain yield most in the no-tillage system.


Introduction
The majority of rice cultivated in the world is under flooded conditions (Prasad, 2011).Upland rice is cultivated in Asia, Africa, and the Americas (Nascente, Crusciol, & Cobucci, 2013).However, as water availability for rice irrigation by flooding decreases, mainly in China and India, upland rice cultivation is increasing in importance (Feng et al., 2007;Prasad, 2011).According to Bouman et al. (2007), irrigated lowland rice receives approximately 24 to 30% of the freshwater consumed worldwide, and these rice systems are threatened because of the increasing demand for freshwater by humans, industry, and animals.Therefore, the development of technologies that conserve water, such as growing rice under aerobic conditions in a no-tillage system (NTS) instead of flooding (Nascente et al., 2013), would augment and stabilize upland rice yields in addition to consuming less water (Crusciol, Soratto, & Mateus, 2006;Heinemann, Stone, & Fageria, 2011).
Upland rice cultivation in an NTS still needs some crop adjustments, such as row spacing and the management of nitrogen fertilization, when aiming for higher grain yield (Carvalho, Soares, & Reis, 2008).Proper row spacing depends on the interactions among cultivars, crop management, and the environment.Upland rice crops under conventional tillage (plowing and disking) is recommended with a range in distance of 0.40-0.50m between rows and a seeding rate of 60 to 70 seeds per meter (Santos, Stone, & Vieira, 2006).However, with the introduction of modern varieties of rice with upright leaves and intermediate height, we have to study the proper row spacing in an NTS because rice plants grown under this soil management typically exhibit slower growth, which encourages weed infestation that negatively affects the grain yield (Nascente et al., 2013).For example, for the cultivar BRSMG Conai, Carvalho et al. (2008) found that increased productivity under an NTS was obtained at row spacing of 0.30 m.In this sense, under an NTS, upland rice cultivars may need smaller row spacing to avoid the infestation of weeds and subsequent reduced productivity and to provide better use of environmental resources (Jadoski, Carlesso, Petry, Woishick, & Cervo, 2000).
The application of larger amounts of nitrogen at sowing time with reduced spacing can perform as a starter and allow an early closed canopy, thus reducing weed infestation (Fageria, 2014).However, if nitrogen is reported to provide significant increases in the productivity of upland rice grain (Arf et al., 2005;Cazetta, Arf, Buzetti, Sá, & Rodrigues, 2008;Cancellier et al., 2011;Nascente et al., 2011), then nitrogen applied at rates higher than recommended could create adequate conditions for the overgrowth of plants and thus increase lodging (Arf et al., 2005;Nascimento et al., 2009;Alvarez, Crusciol, Nascente, Rodrigues, & Habermann, 2012).Furthermore, the excess of N may enable the development of diseases and adversely affect grain yield (Crusciol et al., 2006;Marzari et al., 2007).
The appropriate combination of row spacing with nitrogen rate can provide significant increases in grain yield of upland rice.However, few studies have evaluated these two variables, especially for new rice cultivars with types of intermediate plants (plants with high height and susceptible to lodging when grown on fertile soils or when fertilized with high rates of N) and modern plants (smaller plants and less susceptible to lodging).The objective of this study was to determine the effect of row spacing and nitrogen topdressing fertilization on the plant height, yield components and grain yield, and quality of two new genotypes of upland rice cultivated under no-tillage systems.

Site description
The experiments were conducted at Capivara Farm of the Embrapa Rice and Beans Unit, which is located in Santo Antônio de Goiás, Goiás State, Brazil, at 16º28'00"S and 49º17'00"W and 823 m of elevation.The climate is tropical savanna and is considered Aw according to the Köppen classification.There are two well-defined seasons: a normally dry season from May to September (autumn/winter) and a rainy season from October to April (spring/summer).The average annual rainfall is between 1,500 and 1,700 mm, and the average annual temperature is 22.7°C, ranging annually from 14.2 to 34.8°C.During the period of this study, the temperature and the amount of rainfall data were recorded (Figure 1).
The soil is classified as a clay loam (kaolinitic, thermic Typic Haplorthox) acidic soil.Prior to the study, in 2012 and 2013, chemical analyses were performed in a depth range of 0-0.20 m for the initial characterization of the area (Table 1).Chemical analyses were performed according to the methodology proposed by Claessen (1997).
The experimental area had been cultivated in a crop-livestock integration using a no-tillage system for seven consecutive years, followed by a crop rotation program of soybean (summer), rice (summer) and irrigated common bean (winter), corn + Brachiaria (summer), and two years of grazing pasture.The installation of experiments was conducted in plots where the upland rice was the crop to be grown following the established program of crop rotation.

Rice crop management
Cover crops were desiccated with a glyphosate application (1.8 kg ha -1 acid equivalent) 30 days before sowing of the upland rice.The sowing was performed mechanically using 230 seeds m -2 of rice.The seed was sown on December 2, 2012 and on December 5, 2013.Rice plant emergence occurred five days after sowing for the growing seasons.The base fertilization, to be applied in the sowing furrows, was calculated according to the soil chemical characteristics and the recommendations of Sousa and Lobato (2003).Therefore, sowing fertilization was 15 kg ha -1 of N as urea, 90 kg ha -1 of P 2 O 5 as triple superphosphate, and 45 kg ha -1 of K 2 O, as potassium chloride, in both years.Nitrogen topdressing fertilization (as urea) was done 40 days after the rice emergence, according to each treatment.Cultural practices were performed according to standard recommendations for a rice crop to keep the area free from weeds, diseases, and insects.

Trait Measurements
Rice harvest was carried out by hand after physiological maturity (March 21, 2013 andMarch 25, 2014) of the grain in the usable area of each plot.Plots were evaluated for the plant height (m), which was determined by measuring five plants per plot at the time when the crop was at the phonological stage of pasty grains, with the distance between the soil surface and the top end of the highest panicle being recorded; the number of panicles m -1 , which was determined by counting the number of panicles within 1.0 linear m of one of the rows in the useful area of each plot; the mass of 1000 grains, which was randomly evaluated by collecting and weighing 1000 grains from each plot, corrected to 13% of water content; and the grain yield, which was determined by weighing the harvested grain of each plot, corrected to 13% of the water content and converted to kg ha -1 .
The milling yield was determined only in the 2014/14 growing season by collecting a sample of 100 g of rice grains in each plot, which was processed in a test mill for 1 minute; then, the polished grains were weighed, and the value obtained was regarded as the effective performance, with the results being expressed as a percentage.Subsequently, these polished grains were placed on a grain "Trieur" 2 machine, and the separation of the grains was processed for 30 seconds; the grains remaining in the "Trieur" were weighed to obtain the yield of broken grains, which was also expressed as a percentage.

Statistical analysis
For statistical analysis, the SAS Statistical Software, SAS Institute, Cary, NC, USA (SAS Institute, 1999) was used.In qualitative variables (upland rice yield components, grain yield, milling yield, and broken grains), data were subjected to an analysis of variance, and when the F test proved significant, the data were compared by a Tukey test at p < 0.05.In this case, the factor was row spacing.In the quantitative variables (N rates), results were submitted to regression analysis when p < 0.05.

Row spacing
The row spacing significantly affected the number of tillers, number of panicles, and grain yield of the genotype 07SEQCL441 (Table 2) and the number of tillers, plant height, number of panicles, and grain yield for the genotype BRS Esmeralda (Table 3).In the narrowest spacing (0.225 m), the highest number of tillers (365 tillers m -2 for genotype 07SEQCL441 and 314 tillers m -2 for BRS Esmeralda), plant height (95.5 m for genotype BRS Esmeralda), number of panicles (364 panicles m -2 for genotype 07SEQCL441 and 314 panicles m -2 for BRS Esmeralda), and grain yield (4129 kg ha -1 for genotype 07SEQCL441 and 4,697 kg ha -1 for BRS Esmeralda) were found.

Nitrogen rates
The rate of nitrogen application provided a significant effect on the number of tillers and panicles for both genotypes (Tables 2 and 3), with a quadratic effect (Figures 2 and 3) being observed.Increasing nitrogen rates provided a significant increase in the plant height with the linear fit of the data equation for both genotypes (Figure 4).Nitrogen rates contributed to a decrease in the mass of 1,000 grains for both genotypes (Table 2 and 3), with data fit to quadratic regressions (Figure 5).The genotype 07SEQCL441 CL had a quadratic response with higher yield approximately 50 kg ha -1 of N, whereas the genotype BRS Esmeralda had a linear response, and the rate of 150 kg ha -1 was not enough to allow the total potential of this genotype (Figure 6).The row spacing and nitrogen rates did not significantly affect the milling yield or the broken grains for either genotype (Tables 4 and 5).

Growing seasons
The genotype 07SEQCL441 CL showed higher grain yield in the 2012/13 growing season (Table 2).On the other hand, the genotype BRS Esmeralda yielded more in the 2013/14 growing season (Table 3).No interaction was observed between the row spacing and N rates.Therefore, we discussed the single effects of each factor.

Row spacing
The resulting greater values of the variables in the smallest row spacing may reflect the greater competition among plants in the larger row spacing (0.45 m), as the population per square meter for both genotypes was the same in all row spacing (230 seeds m -2 ).Therefore, the greater row spacing (0.45 m) had the highest number of plants per row and provided more competition among them.On the other hand, in the smallest row spacing (0.225 m), a better distribution of the plants occurred in the area.According to Jadoski et al. (2000) and Freitas, Nascente, and Santos (2013), a larger number of plants provides increased competition for light and photoassimilates.According to Sangoi (2001), plants equidistantly spaced compete minimally for nutrients, light, and other factors.Jadoski et al. (2000) also observed that reduced spacing, within certain limits, provides a more equidistant distribution between plants to reduce the competition for environmental resources, thus more completely and efficiently exploiting a given area of N rates (Kg há -1 ) N rates (Kg há -1 ) N rates (Kg há -1 ) N rates (Kg há -1 ) soil.Therefore, we can infer that for these genotypes (07SEQCL441 CL and BRS Esmeralda) the row spacing of 0.225 m provided better distribution of the plants in the area and allowed higher grain yield than the other row spacing (0.45 and 0.35 m).Furthermore, the low plant competition in the narrowest row spacing (0.225 m) in comparison to the other row spacings, allows an increase in the number of tillers and, as a result, the number of panicles per area and grain yield.Freitas et al. (2013) reported that larger row spacing entails the reduction in the solar radiation trapping efficiency per area and can provide increased yield per plant; however, a reduction in the productivity per area also occurs.Conversely, the smallest row spacing provides a higher yield per unit area and reductions in the productivity per individual.We provided proper weed control, which was not tested in this research, and therefore, an evaluation of the influence of row spacing on weed infestation was not The rice grain yield is determined by three components: the number of panicles m -2 , number of grain spikelets per panicle, and mass of 1,000 grains (Yoshida, 1981).Thus, from the results obtained of the higher number of panicles m -2 , rice cultivation in the row spacing of 0.225 m was expected to provide the highest yield, which was confirmed, as the highest values were observed (4,129 kg ha -1 for genotype 07SEQCL441 CL and 4,697 kg ha -1 for genotype BRS Esmeralda) under this condition (Table 3).Carvalho et al. (2008) also observed higher grain yield in the cultivar BRSMG Conai at the narrowest row spacing (0.30 m) compared to the greatest row spacing (0.40 m).

Nitrogen rates
Nitrogen had a marked effect on the number of tillers and panicles for both genotypes.According to Larrosa, Felipe, Marchezan, Aita, and Coradini (2001), nitrogen application increases the number of tillers and panicles.Nitrogen is essential for cell structures and functions, for all enzymatic reactions in the plant, and for part of the chlorophyll molecule (photosynthesis), among other functions (Fageria, 2014).Therefore, increasing the N rates likely provided better conditions for plant growth and improved the number of tillers and panicles per area.
According to Arf et al. (2005), nitrogen is the nutrient that most affects plant height.Corroborating this information, Buzetti et al. (2006); Hernandes, Buzetti, Andreotti, Arf, and Sá (2010); and Mattje, Fidelis, Aguiar, Brandão, and Santos (2013) also reported significant increases in the height of rice plants with increasing rates of N. Despite this increase in plant height, no lodging was observed.Nitrogen can make plants more susceptible to lodging, as reported by Fidelis et al. (2012).However, the genotypes used showed great resistance to lodging under high rates of nitrogen application.Moreover, under the NTS, plants grow less than with conventional tillage (Nascente et al., 2013), and in this study, the plants probably did not grow high enough to cause lodging.
The nitrogen rates provided decreases in the mass of 1,000 grains for both genotypes.This may have occurred because, when the number of panicles increases, a tendency occurs for the mass of grains to decrease (Santos et al., 2006).The grain yield showed a positive effect from increasing N rates.Increased N rates typically provide significant increases in the productivity of rice grains, as reported by several authors (Arf et al., 2005;Cazetta et al., 2008;Cancellier et al., 2011;Nascente et al., 2011).The current results show that the genotypes responded differently to the nitrogen rates.Lower rates of N were required for 07SEQCL441 CL to achieve high yield (approximately 3,900 kg ha -1 ) compared to BRS Esmeralda, which required higher rates of N (150 kg ha -1 ) to achieve higher grain yield (>4,500 kg ha -1 ).
In both years, the rice grain yield without N application was high (3,659 and 3,947 kg ha -1 for the growing season 2012/13 and 2013/14, respectively).This yield could be attributable to the high soil fertility of the places where the trials were performed (Table 1, Sousa & Lobato, 2003).According to Santos et al. (2006), in soils with high fertility and adequate soil moisture (Figure 1), rice can achieve high grain yields, as observed in this study.Furthermore, the high soil fertility resulted from the use of the areas for crop-livestock rotation, where an input of soil organic matter occurs every year.
The row spacing and nitrogen rates did not significantly affect the milling yield and broken grains for both genotypes.This is a preferred result, as the presence of broken grains on a plot of rice is an undesirable trait because it reduces the quality and commercial value of the product.In addition to reducing the economic value, grain fractions can be eliminated along with the shells, thereby decreasing the total amount of husked grains and the yield (Crusciol et al., 2006).These authors also reported that although the causes that determine the quality of the grain are not well understood, inadequate crop management is known to have a direct influence on this attribute.Thus, it appears that neither the spacing nor the N levels affect the quality of rice grain.Neves, Buzetti, Arf, and Sá (2004); Marzari et al. (2007); and Artigiani, Crusciol, Arf, Alvarez, and Nascente (2012) also reported no effect of the N application rate on the grain quality of rice.In addition, Santos, Castro, Soares, and Cornélio (2002) reported no response in the quality of upland rice grain due to row spacing.

Growing seasons
The higher grain yield achieved by the genotype 07SEQCL441 CL in the 2012/2013 growing season could be explained by the higher number of panicles m -2 and grain mass obtained in this growing season.On the other hand, for the genotype BRS Esmeralda, which yielded more in the growing 2013/2014 season, the higher grain yield could be due to the greater mass of 1,000 grains in the growing season 2013/14 compared to 2012/13.Regarding rainy conditions, in both years, the amount of rain and its distribution was appropriate to the development of the upland rice.According to Reichardt (1987), rice plants need from 450 to 700 mm of rain, during the crop cycle.In our trials, 937.4 and 1044.8 mm of rain was recorded for the 2012/2013 and 2013/2014 growing seasons, respectively.In addition, Rodrigues, Soratto, and Arf (2004) suggested that the rainfall should be well distributed during the crop cycle, and such distributions of rain were observed in both years of the current study (Figure 1).

Conclusion
The 0.225 m row spacing provided the highest yield of the upland rice grain, and the nitrogen application rates allowed increasing yield of the upland rice grain for both genotypes.However, for the genotype 07SEQCL441 CL, the yield increased only until reaching the rate of 50 kg N ha -1 .Therefore, row spacing and nitrogen rates do not affect the grain quality of rice.

Figure 1 .
Figure 1.Maximum (T maximum), minimum (T minimum) and average (T average) temperatures and rainfall during the trial period of upland rice grown under a no-tillage system in the experimental fields of Santo Antônio de Goiás in the 2012/13 and 2013/14 growing seasons.

Figure 5 .Figure 6 .
Figure 5. Mass of 1000 grains (g) of rice plants of the genotypes 07SEQCL441 CL and BRS Esmeralda in a no-tillage system as a function of the nitrogen rate.Santo Antônio de Goiás, Brazil.Growing seasons 2012/13 and 2013/14.

Table 1 .
Chemical characteristics of soil in the experimental areas.Growing seasons 2012/13 and 2013/14.

Table 2 .
Plant height (PH), number of tillers (NT), number of panicles m -2 (PAN), mass of 100 grains (MGRAIN), and grain yield (YIELD) of the upland rice genotype 07SEQCL441 CL in a no-tillage system as a function of row spacing and nitrogen rates.Santo Antônio de Goiás, Brazil.Growing seasons 2012/13 and 2013/14.

Table 3 .
Plant height (PH), number of tillers (NT), number of panicles m -2 (PAN), mass of 100 grains (MGRAIN), and grain yield (YIELD) of upland rice genotype BRS Esmeralda in a no-tillage system as a function of row spacing and nitrogen rates.Santo Antônio de Goiás, Brazil.Growing seasons 2012/13 and 2013/14.Number of panicles of the rice plants of the genotypes 07SEQCL441 CL and BRS Esmeralda in a no-tillage system as a function of the nitrogen rate.Santo Antônio de Goiás, Brazil.Growing season 2013/14.Plant height (cm) of rice plants of the genotypes 07SEQCL441 CL and BRS Esmeralda in a no-tillage system as a function of the nitrogen rate.Santo Antônio de Goiás, Brazil.Growing seasons 2012/13 and 2013/14.
¹Means followed by the same letter in a column do not differ by Tukey's test for p ≤ 0.05. Figure 2. Number of tillers of rice plants of the genotypes 07SEQCL441 CL and BRS Esmeralda in a no-tillage system as a function of the nitrogen rate.Santo Antônio de Goiás, Brazil.Growing seasons 2012/13 and 2013/14.Acta Scientiarum.Agronomy Maringá, v. 38, n. 4, p. 493-502, Oct.-Dec., 2016

Table 5 .
Milling yield (MY) and broken grains (BG) of rice plants of the genotype BRS Esmeralda in a no-tillage system as a function of the row spacing and nitrogen rates.Santo Antônio de Goiás, Brazil.Growing season 2013/14.