The management of sowing density on yield and lodging in the main oat biotype grown in Brazil

1 Department of Physical and Engineering Sciences, Regional University of the Northwest of Rio Grande do Sul, 480 Lulu Ingelfritz Street, Zip code: 98700-000, Ijuí, RS, Brazil. 2 Department of Agrarian Studies, Regional University of the Northwest of Rio Grande do Sul, 3000 Comércio Street, Zip code: 98700-000, Ijuí, RS, Brazil. 3 Department of Crop Plants, Federal University of Rio Grande do Sul, 7712 Bento Gonçalves Avenue, Zip code: 91540-000, Porto Alegre, RS, Brazil.


INTRODUCTION
Oats (Avena sativa L.) is considered a multi-purpose cereal and its grains have excellent nutritional value for human and animal consumption (Hawerroth et al., 2013).The greatest expression of oat productivity is directly associated with the management techniques, such as the nutrients availability, plant population, phytosanitary control, among others (Benin et al., 2005;Silva et al., 2015).The plant population is an important factor in the potential expression of biomass and grains yield in cereals (Ceccon et al., 2004), its variation is associated with the genotype potential in producing fertile tillers, once the sowing density influences the number of spikes and/or panicles produced by area (Sparkes et al., 2006;Valério et al., 2009).The increase in the number of tillers and/or plants per area acts directly on the vegetal biomass, an important aspect to enhance biological productivity and the relationship between the straw and grain mass (Ozturk et al., 2006;Silveira et al., 2010).It is highlighted that the rapid coverage of vegetal biomass on the soil by the canopy adjustment may favor a better use of light and nutrients for grain yield and a more effective control in the weeds evolution (Fleck et al., 2009;Silva et al., 2012).
The technical specifications of the Brazilian Commission of Oat Research have suggested sowing density in 200 to 300 seeds m -2 , a condition adopted since that the cultivation of this species began to have commercial importance in the 90s.However, the continuous oat breeding has modified the plant architecture, among other features, changing the high stature biotype, late cycle and high relation straw/grain, for genotypes with stature lower than one meter, reduced cycle and greater caryopsis volume in relation to husk (Hawerroth et al., 2015;Silva et al., 2015).Therefore, changes that may modify the response of cultivars to plant population, suggest the need for more adjusted recommendations to the actual biotype of white oats grown in the Southern Brazil.
In agricultural systems, the type of biomass on soil influences the dynamics of release and use of N-residual in the expression of the yield components (Mantai et al., 2015).The wheat and oat grown on the soybeans and corn residue show differences in the tillers and canopy development, reflecting directly on biomass and grain yield (Wendling et al., 2007;Mantai et al., 2015).Aside from this, the constant climate changes has also changed the vegetal productivity, demonstrating the necessity of more stressed tolerant plant varieties (Araus et al., 2008) and efficient in the use of light and nutrients (Oliveira et al., 2011;Costa et al., 2013).Therefore, the oat productivity is directly associated to the use of fertilizers, cultivation techniques and soil and edaphoclimatic conditions (Costa et al., 2013;Silva et al., 2015).Although the favorable cultivation conditions may increase vegetal productivity, they tend to promote increased vegetative vigor, making favorable the lodging occurrence, condition that brings serious losses in yield and grains quality (Silva et al., 2012).
The occurrence of rains, winds and soil condition with a higher nitrogen content, may increase the occurrence of plant lodging, whose conditions are common variables along the crops cultivation (Berry et al., 2003).In this context, the use of sustainable technologies and low cost, such as vegetation cover management, the use of the Nresidual and the density cultivation adjustment on the main oat biotype grown in the Southern Brazil, can bring benefits to maximize the biomass and grain yield with lodging reduction.
The aim of this study is the sowing density adjustment of early cycle's oat cultivars and reduced stature to increase biomass and grain yield with reduced lodging, considering high and low N-residual release succession systems in different years.

MATERIALS AND METHODS
The study was developed in the field during the years 2011, 2012 and 2013 in Augusto Pestana city, RS state, Brazil (28°26'30'' South latitude and 54°00'58'' West longitude).The soil of the experimental area is classified as Distrofic Red Latosol Typical, which its U.S. equivalent is Rhodic Hapludox (USDA, 2014), and the climate of the region, according to Köppen classification, is 'Cfa type', with hot summer without a dry season.In the study, ten days before sowing, soil analysis was performed and it was identified in the following chemical characteristics of the local: i) corn/oat system (pH = 6.5, P = 34.4mg dm -3 , K = 262 mg dm -3 , Organic matter = 3.5%, Al = 0.0 cmolc dm -3 , Ca = 6.6 cmolc dm -3 and Mg = 3.4 cmolc dm -3 ) and ii) soybean/oat system (pH = 6.2, P = 33.9mg dm -3 , K = 200 mg dm -3 , Organic matter = 3.4%, Al = 0.0 cmolc dm -3 , Ca = 6.5 cmolc dm -3 and Mg = 2.5 cmolc dm -3 ).During the three years, sowing was performed in the second fortnight of May with seederfertilizer for composition of 5 rows of 5 m in length and row spacing of 0.20 m, forming the experimental unit of 5 m 2 .During the study execution, tebuconazole fungicide applications were made at the dosage of 0.75 L ha -1 .Moreover, the weeds control was carried out with metsulfuron-methyl herbicide at a dose of 4 g ha -1 and additional weeding whenever necessary.At the oats sowing time 80 and 60 kg ha -1 of P2O5 and K2O, were applied, respectively, based on levels of P and K in the soil and nitrogen base with 10 kg ha -1 and rest of N applied in topdressing on the phenological stage of fourth leaf expanded, expecting thus grain yield about 3 t ha -1 .The studies were carried out considering the two main succession systems used in southern Brazil for oats, involving soil coverage with vegetable residue of high and reduced carbon/nitrogen ratio in the corn/oat and soybean/oat succession systems, respectively.In each succession system two experiments were conducted, one to quantify the biomass production rate by the cuts made every 30 days until the harvesting point, and the other, for analysis of plant lodging and grain harvest to estimate yield.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License cultivars used represent current genotypes with biotype desired in southern Brazil, with similarity to the cycle (early), height (reduced) and lodging (moderately resistant), however, distinguished in production tillering capacity (Brisasul = high; URS Taura = reduced).
Grain yield was obtained by cutting three central rows of each plot in the moment defined as the last cut in the experiment directed to analysis of biomass production rate (120 days), stage near the harvesting point, with grain moisture around 15%.The plants were threshed with a stationary harvester and directed the laboratory to correct grain moisture to 13%, and weighing to estimate grain yield (GY, kg ha -1 ).In experiments aiming to quantify the biomass production rate in each succession system, the harvest of the plant material was held close to the ground at the 30, 60, 90 and 120 days after the emergency, totaling four cuts through collection of a linear meter of three central rows of each plot.The green biomass samples were directed to forced-air oven at a temperature 65°C, until it reached constant weight to estimation of total dry matter produced (TDM, kg ha -1 ).
The lodging index was estimated visually and expressed in percentage, having considered the angle formed in the vertical position of the plants culm in relation to the ground and the area of lodged plants.For this estimate it was used the methodology suggested by Moes and Stobbe (1991), modified, with the lodging index (LODG-I) defined by the equation: LODG-I% = I × LODG × 2; where (I) reflects the plants inclination degree, ranging from 0 to 5 (0, absence of inclination and 5, all plants completely lodged); LODG represents the area with lodged plants in the plot, which ranges from 0 to 10, where 0 corresponds to the absence of lodged plants in the plot and 10 to lodged plants over the whole plot, regardless of their inclination.Therefore, this equation considered the incidence and severity of plants lodging, for example, when there is I = 5 and LODG = 10, LODG-I%= 5×10×2 = 100%, which corresponds to the existence of lodged plants close to the ground in the total area of plot.
To meet the homogeneity and normality assumptions via Bartlett tests, analysis of variance were performed for the detection of the main effects and interaction.Based on this information, it was proceeded the linear equation adjustment  in estimating the biomass production day -1 ha -1 rate and the averages comparison by Scott and Knott ( 1974) at the analysis of points of density considered on grain yield and lodging.Afterwards, with the equation adjustment of degree two grain yield (GY), it was obtained the optimal sowing density (X) by the equation . It was carried out regression equation adjustment (linear or quadratic) that describes the behavior expression of percentage of oat plant lodging by increasing the cultivation density.In these equations, oat lodging estimate was performed in distinct cultivation conditions using optimal sowing density (X) at maximum grain yield.For all the determinations employed, computational program GENES (Cruz, 2013) was used.It is highlighted that the average grain yield values per crop year along with the temperature and rainfall information were used to classify the years as favorable and unfavorable.

RESULTS AND DISCUSSION
In this study, significant differences were detected between the main effects and interaction between Romitti et al. 1937 cultivars, years and sowing densities.Therefore, the results are presented in order to unfold the effects of this interaction.In Table 1, about the soybean/oat succession system, the highest rate of biomass production day -1 was obtained at a density of 900 seeds m -2 , regardless of the years and cultivars evaluated.The increase of sowing density in oat indicates the biomass rate fostering; however, the most expressive average values of grain yield were obtained at the points of 300 and/or 600 seed m -2 . In these two densities, the 600 seeds m -2 provided in most cases, greater plant lodging.However, the highest biomass day -1 rate for rapid ground cover in the use of light and weed control with grain yield, is more efficient at the point of 600 seeds m -2 . Although this density facilitates greater plant lodging in this system in relation to 300 seeds m -2 , the tendency of increased lodging can be circumvented by the use of growth reducers, strategy that has been adopted for oats and other cereals (Hawerroth et al., 2015).
In Table 1, the rapid N-residual release soybean/oat system, the biomass production day -1 rate and grain yield indicated in general terms, cultivar URS-Taura is larger than Brisasul and with greater stability of the grain yield and lodging in different densities tested, regardless of cultivation year.It is highlighted, in this condition, that the most favorable years to increase of grain yield were also those that provided greater plant lodging.The years 2011 and 2013 indicated the most expressive average values (Table 3) as a result of climate conditions temperature and more favorable rainfall for the cultivation, mainly in 2013, with mild temperatures and adequate rain distribution during the crop cycle (Figure 1).It is verified that in a condition that high biomass and grain yield are searched by means of combined use of N-fertilizer and N-residual, the possibility of plant lodging becomes stronger on condition of favorable cultivation year.
In Table 2, corn/oat system, the equations obtained also indicated higher biomass production day -1 rate in higher density, justifying the fact that they are positively related.In this condition, the highest yields were found at 300 and 600 seeds m -2 .In the analysis of the biomass production day -1 rate with grain yield, regardless of year and cultivar, the point of 600 seeds m -2 also appears as more indicated.In this condition of high C/N ratio (corn/oat system), the low N-residual release rate seems to influence on sowing density when compared to soybean/oat system, because in most situations, the maximum grain yield was obtained at the point of 600 seeds m -2 . The year 2011 stands out when the cultivar URS Taura also indicated maximum yields with 900 seeds m -2 . In this way, differences between cultivars were detected, with cultivar URS Taura showing in 2011 and 2012 the greatest grain yield, but in 2013, the most expressive results were obtained with the cultivar Brisasul.A relevant fact was the reduced lodging observed in all conditions tested, reporting that the corn/oat system proves efficient in reducing plant lodging (Table 2).Although in this system, the N-residual rate release is reduced, benefits in the tissues structure was obtained, possibly by increasing lignin content, generating greater culm resistance.
Concerning the oats aimed at human food, this condition shows relevant, because the value of the product for the industry is reflected in high mass of grains and the hectoliter, strongly damaged traits in occurrence of lodging.
The determination of the biomass production rate is decisive in determining the sowing density adjustment, and it allows the establishment that will favor better use of light and nutrients in association with natural weeds control (Fleck et al., 2009;Silva et al., 2012).However, the use of very elevated densities may reduce the grain yield through strong intraspecific competition and encourage plant lodging (Krüger et al., 2011;Hawerroth et al., 2015).Pinthus (1973) defined the lodging as a permanent modification state of the culm position in relation to its original position, resulting in curved plants or broken culms.The breaking and/or lodging are complex phenomena, and their expression depends on genetic factors interrelated with the climate, soil and management practices (Fontoura et al., 2006).Among the main agents that promote the plant lodging, the wind and rain stand out (Easson et al., 1993;Silva et al., 2006).In favoring conditions to lodging, the use of plant growth regulator has been used as an efficient solution, including allowing the use of higher sowing densities and larger doses of nitrogen in increasing yield (Matysiak, 2006;Schwerz et al., 2015).Rademacher (2000) defines plant growth regulator as a synthetic compound capable of reducing undesirable longitudinal growth of the aerial part of plants, with no decrease in grain yield.
In the definition of the optimal sowing density in oats and their reflexes about the plant lodging, in Tables 3  and 4 equations presented that the grain yield behavior and the lodging, validated by the analysis of variance of regression (not displayed) and test on the inclination parameter (b ix n ) that were significant.In Table 3, in the soybean/oat system, regardless of year and cultivar, the grain yield showed quadratic trend and the lodging linear adjustment at the sowing densities increase, except in 2012, because the cultivars showed quadratic behavior in plant lodging.In the years 2011 and 2012 in soybean/oat system Table 3), independent of the cultivars tested, the optimal density cultivation for grain yield was obtained Table 3. Regression equation to estimate the optimal density on grain yield and reflections on the plant lodging in soybean/oat system.

Cultivar
Equation: y= a±b1x±b with values greater than 500 seeds m -2 . On the other hand, in 2013, the adjusted density results indicated values below 500 seeds m -2 , but higher than the recommendation of the species which is 200 to 300 seeds m -2 .The equations obtained with the use of optimal sowing density indicated grain yield estimates of more than 3 t ha -1 at the soybean/oat system, in some situations, surpassing 4 t ha -1 (Table 3).In this way, the optimal sowing density for maximum grain yield to be included in the model that describes the lodging in oats, allowed estimation of lodged plants by using the optimal density.Therefore, in favorable cultivation years (2011 and 2013), the lodging was estimated between 30 and 48%, indicating that the adjusted density in soybean/oat system can bring losses by the plants falling and compromising the grain quality.It is highlighted that the genotypes tested besides showing cycle and reduced height are also described as moderately resistant to plant lodging (Silva et al., 2015).In unfavorable cultivation year (2012), the plant lodging was almost nonexistent, however, with the lowest grain yield (Table 3).
The favorable cultivation conditions defined by temperature and rainfall (Figure 1) were decisive on the plant lodging.Therefore, the improvement of investments with fertilization in favorable years can step up lodging, especially with the use of higher sowing density seeking maximum yield.The general analysis (Table 3), regardless of the cultivars and cultivation years, the density adjusted to maximum yield in the soybean/oat system was around 500 seeds m -2 (x = 506), with 40% of lodging.
In Table 4, corn/oats succession system, behavior similar to soybean/oat system were observed, with adjustment of quadratic functions in the estimation of grain yield and linearity on the plant lodging in the increased density of sowing, regardless of year and cultivar.In this system, the adjusted densities for maximum grain yield was higher than 500 seeds m -2 , however, situations such as in 2011, which showed optimal density around 600 seeds m -2 (x = 618) on the cultivar URS Taura.This condition reinforces the hypothesis of dependence of the sowing density adjustment by succession system and cultivar, mainly influenced by genetic differences of the lowest production rate and development of tillers observed in cultivar URS Taura, and intensified on condition of lower N-residual availability.In this system, grain yield above 3 t ha -1 were also obtained, highlighting the cultivar Brisasul in more favorable cultivation year (2013) with 3.6 t ha -1 .In the behavior linear equation of plant lodging, the inclusion of the optimal density for grain yield showed lodging estimates totally different from those obtained in the soybean/oat system.The use of optimal sowing density on this system, independent of favorable and unfavorable year of cultivation and genetic differences between cultivars, allowed efficient reduction in plant lodging.Therefore, management condition evidenced the increase in grain yield with the use of optimal density without lodging.This fact reports the possibility of greater ease of harvest and maintenance of the physical and chemical qualities of the grain, characteristic strongly required by industry.Valério et al. (2008) observed that wheat genotypes with reduced tillering are more dependent on the sowing density increase.The genotypes of high tillering submitted to high densities, suffer greater competition for water, light and nutrients, reducing the grain yield and favoring the plant lodging (Ozturk et al., 2006).Therefore, the identification of a stable sowing density and responsive to cultivation improvements can foster greater grain yield in the ideal balance of development of yield components and with less risk in the plants fall (Silva et al., 2012).In wheat, it was observed that the equidistant distribution of seeds at density of 350 to 500 seeds m -2 , promoted greater grain yield in the cultivars evaluated (Silveira et al., 2010).They also observed that the sowing density adjustment was changed by genotype and cultivation year, with yield increase with stability in higher density.Mantai et al. (2015) observed that the type of vegetal residue by the succession system in oat significantly alters the expression of biomass and grain yield.According to these authors, the N-fertilizer, although being used in less proportion in the soybean/oat system, brings greater plant lodging risks, especially in favorable.
On the other hand, increased doses of N-fertilizer did not promote plant lodging in the corn/oat system, which cultivation conditions reinforces the results obtained in this study, by the contribution of this system in reducing plants fall.Abreu et al. (2005) by studying delayed cycle oat genotypes observed linear increase in biomass and grain yield by increasing the population from 100 to 400 plants m -2 . Silva et al. (2015) highlight that the proposal of recommendation of seeds higher than the techniques currently used in oats can increase grain yield provided that there is no lodging, and with benefits in the management of the crop by the greater vegetation cover, be through a more effective weeds control or through moisture maintenance and soil erosion control, qualifying the direct sowing system for the summer species.

Conclusions
Regardless of year and cultivar, the main oat biotype grown in Southern Brazil shows optimal sowing density around 500 seeds m -2 in the biomass and grain yield expression in the corn/oat and soybean/oat succession systems.
In the soybean/oat succession system, the use of optimal density can promote the plant lodging, mainly in favorable cultivation years.
The corn/oat succession system although be of lower N-residual release, proves efficient the grain yield expression and reduced lodging in the use of adjusted density.

Figure 1 .
Figure 1.Different years of study climatological data.

Table 1 .
Regression equation of total dry matter and grain yield averages and lodging oat cultivars in sowing densities on soybean/oat system.

Table 2 .
Regression equation of total dry matter and grain yield averages and lodging oat cultivars in sowing densities in corn/oat system.

Table 4 .
Regression equation to estimate the optimal density on grain yield and reflections on the plant lodging in corn/oat system.