Wheat cultivars under bulk density levels in Cerrado Rhodic Hapludox , Central Brazil

The Brazilian Cerrado presents potential to expand its wheat cultivation, but soil compaction is one of the factors that may limit production. The objective of this study was to evaluate production characteristics and chlorophyll index of two wheat cultivars (Triticum aestivum L.) under bulk density levels. The experiment was conducted in a greenhouse, with Rhodic Hapludox collected at 0.00-0.20 m depth. The experimental design was entirely randomized and the treatments were arranged in a 5x2 factorial scheme, corresponding to five bulk densities levels (1.0; 1.2; 1.4; 1.6 and 1.8 Mg m -3 ) and two wheat cultivars (BRS Guamirim and IAC 350) with four replications. The experimental plot was composed of one poly (vinyl chloride) pot of 0.1913 m internal diameter and 0.20 m height. Variables evaluated were: spike number, dry mass of spikes, aerial part, roots and chlorophyll index. There was no significant interaction between wheat cultivars and bulk density levels. The increasing bulk density reduced production and chlorophyll index of the wheat cultivars BRS Guamirim and IAC 350. The IAC 350 cultivar presented better spike production and higher chlorophyll index, regardless the bulk density levels.


INTRODUCTION
Brazil's greatest wheat production is located in the southern region of the county, which is responsible for more than 90% of the national production (Conab, 2012).However, the central-western region is also a very good alternative for the wheat production expansion under both non-and irrigated conditions (Coelho et al., 2010).Wheat flour is the main raw material used for food preparation, such as bread, biscuits, cakes and pastries.Therefore, improvement of wheat production potential via soil management practices is one of the main research challenges facing the worldwide growing demand for food.In this sense, concerns on soil compaction are important, as soils may have appropriate nutrient contents, but because of physical conditions, efficiency of nutrient absorption by plants may be affected, resulting in yield decrease (Bonelli et al., 2011;Cabral et al., 2012).
The soil compaction results in a rearrangement of particles and aggregates due to pressure application on *Corresponding author.E-mail: embonfim@hotmail.comsoil surface by intense traffic of machinery andimplements used in soil management practices (Hakansson and Lipiec, 2000;Reichert et al., 2009).A direct compaction consequence is the elimination of existing gases within macropores, reducing porous space and increasing density.This results in a mechanical barrier to root growth, impairing water and nutrient absorption (Batey, 2009), especially of nitrogen, which is absorbed by mass flow.Also, it negatively influences gas exchange in the soil-plant system, reducing yield.
Soil compaction issues became extremely worrying due to intensification of agricultural mechanization with the indiscriminate use of heavy machines under conditions of high soil moisture (Roboredo et al., 2010).Secco et al. (2009), who studied the compaction effect on no-till Rhodic Hapludox in the Brazilian southern region, verified that high bulk density values (1.62 and 1.54 Mg m -3 ) promoted decreases in wheat yields that ranged from 18.35 to 34.05%.
Currently, information on the bulk density levels that may restrict crop yields in the Brazilian Cerrado is scarce, especially on wheat.In this context, basic researches need to be developed with the aim to evaluate: bulk density levels that may compromise the wheat production potential; and also, whether there are adaptation differences to the bulk density levels among the wheat cultivars recommended for Cerrado soil and climate conditions.Results from this study may be used as a basis for future field researches on soil compaction for wheat crops.
The objective of this study was to evaluate the effect of Rhodic Hapludox bulk density levels on production characteristics and chlorophyll index of two wheat cultivars (BRS Guamirim and IAC 350) grown in a greenhouse in the Cerrado of Mato Grosso State, Brazil.

MATERIALS AND METHODS
The experiment was conducted in a greenhouse located in Rondonópolis, Mato Grosso State, Brazil, with geographic coordinates of 16°27'49" S and 50°34'47" W, Figure 1, from December 2011 to January 2012.The used soil collected from 0.00-0.20 m depth, was classified as Rhodic Hapludox (Santos et al., 2006); with the following chemical and textural characteristics: pH 4.1 (CaCl2); 2.4 mg dm -3 P; 28 mg dm -3 K; 0.3 cmolc dm -3 Ca; 0.2 cmolc dm -3 Mg; 4.2 cmolc dm -3 H; 1.1 cmolc dm -3 Al; 5.9 cmolc dm -3 CEC; base saturation of 9.8 (V%); 22.7 g dm -3 OM; 549 g kg -1 sand; 84 g kg -1 silt; and 367 g kg -1 clay (Claessen et al., 1997).For experiment implementation in the greenhouse, the soil sample was sieved on a 4 mm mesh (Silva et al., 2006) to remove root fragments or particles larger than that diameter that may exist in the soil.Soil acidity was corrected with incorporation of dolomitic limestone (80.3% relative power of total neutralization) into soil samples of 8 dm 3 , increasing the base saturation to 60%.After liming, soil samples were moistened at 80% water retention capacity and placed in plastic bags for 30 days (soil incubation period with limestone).
After incubation with limestone for soil acidity correction, basic fertilization was performed with incorporation of solid granular fertilizer, using 100 mg dm -3 N, 300 mg dm -3 P2O5, and 150 mg dm -3 K2O; fertilizer sources were urea, superphosphate and potassium chloride respectively.The pot entire soil volume was homogeneously fertilized, ensuring the same fertilization level for all    The experimental unit was represented by one poly (vinyl chloride) pot of 0.1913 m internal diameter and 0.20 m height, comprising 5.748 dm 3 .Each pot was composed of two cylinders of 0.10 m height, which were joined by a duct tape; the anti-aphid screen was used to close the pot bases which were affixed with a rubber ring, Figure 2.
Bulk density was used as a parameter to represent soil compaction levels (Grossman and Reinsch, 2002).For all bulk density levels, the compacted layer thickness was 0.05 m, comprising 1.44 dm 3 of the total pot volume, which was 5.748 dm 3 , Figure 3. Dry soil masses varied according to the bulk density level of each treatment.Therefore, Equation 1 was used to determine the dry soil mass to compose the 0.05 m compacted layer in the pots with internal diameter of 0.1913 m (volume of compacted layer = 1.44 dm 3 ).Dry soil masses were 1.44; 1.73; 2.02; 2.30 and 2.59 kg for the respective bulk densities 1.0; 1.2; 1.6 and 1.8 Mg m -3 . (1) Where: Bd -Bulk density (Mg m -3 ) Dm -nDry soil mass (Mg) V -Cylinder volume corresponding to the 0.05 m layer (m 3 ).
According to the Proctor test (Abnt, 1986) 16% moisture was adopted as compaction optimum moisture for the intermediate bulk density level which was 1.4 Mg m -3 .The mean moisture of 16% is used in the laboratory of soil physics where compaction tests, as pilot tests, were performed for the study.To achieve the bulk density levels of 1.0; 1.2; 1.6 and 1.8 Mg m -3 for 16% soil moisture, the compaction energy applied by the press varied, so that curve construction for determination of the ideal moisture for each bulk density was not necessary.
Moisture of the soil reserved for this experiment was determined by the gravimetric method (Claessen et al., 1997) using Equation 2. The mean moisture value found for soil samples was 16%.Determination of the wet soil mass that would be compacted from the dry soil mass corresponding to each treatment (bulk density) was performed according to Equation 3, as well as determination of the sample water content. (2) (3) (2) (2) (3) (3) Where: Wm -Wet soil mass (g) Dm -Dry soil mass (g) Sm -Soil moisture-based mass (g g -1 ) For pot implementation (experimental units), half of the upper cylinder was filled with moist soil (16%) according to the predetermined masses from Equation 1.
The compacted layers were performed with the aim of a hydraulic press with pressure capacity of 15 tons (BOVENAU ® brand, model P15ST).Thickness of the compacted layer for all experimental units was 0.05 m that corresponded to 1.44 dm³ volume.A marked wooden mold was used to indicate the moment the compacted layer achieved 0.05 m thickness, indicating when to stop applying pressure, Figure 4.
The lower cylinder was completely filled with non-compacted soil (density of 1.0 Mg m -3 ).The upper one was filled with the previously compacted layer and placed on top of the lower cylinder.Both cylinders were joined with duct tape.The upper one was then completed with soil at the density of 1.0 Mg m -3 above the compacted layer (1.44 dm 3 ), Figure 5. Plastic trays were used at the pot bottoms to aim at the irrigation by capillarity.
The experimental design was entirely randomized and treatments were arranged in a 5x2 factorial scheme, with five bulk density levels (1.0; 1.2; 1.4; 1.6 and 1.8 Mg m -3 ); two wheat cultivars (BRS Guamirim and IAC 350) and four replications.Twenty seeds were sown per pot.At 12 days after sowing, nitrogen was applied, using urea as the source at the dose of 100 mg dm -3 .Seedling thinning was performed at 15 days after sowing, remaining 5 wheat plants per pot which were cultivated for 54 days.Controlled surface irrigation were carried out until plant establishment (20 days after sowing), then soil moisture was maintained according to the proposed methodology by Silva et al. (2006).After the cultivation period (54 days), spike number and dry mass of spikes, aerial part and roots were evaluated.Also, SPAD readings were performed for determination of the chlorophyll index using the chlorophyll meter Minolta SPAD-502, which was fixed at the middle third of two leaves after the flag leaf (leaves +1 and +2).The mean of 10 readings was considered the SPAD value for each experimental unit.Spike number from each pot was evaluated before harvest.The aerial part was then cut close to the soil surface and leaf mass was separated from spikes.Roots were also collected and washed on a 4-mm sieve.All collected material was placed in paper bags, dried in an airtight circulation heater at 65 °C for 72 h and weighed.
Data were subjected to variance analysis (F test); when significant mean of the wheat cultivars were compared by the Tukey test, while the bulk density levels were submitted to regression analysis, both at 5% probability (p<0.05), using SISVAR 5.3 software (Ferreira, 2008).

RESULTS AND DISCUSSION
Number and dry mass of spikes presented isolated effects for wheat cultivars and bulk density shown in Table 1 and Figure 6.Regardless of the bulk density levels, the IAC 350 cultivar, when compared with BRS Guamirim, had higher number and production of spikes, Table 1.
Considering that a crop production potential is linked to nutrient absorption, this result indicates that IAC 350 has greater capacity to absorb nutrients regardless of the studied bulk density levels.Also, this result may be assigned to the cultivar intrinsic genetic factors, since there was no interaction among bulk density and wheat cultivars.Similarly, Fageria et al. (1995), when studying the response of rice genotypes to soil fertility, reported that absorption and appropriate nutrient use by rice plants are subjected to physiological processes inherent to the studied cultivars.
For bulk density, number and dry mass of spikes were described by regression quadratic models shown in Figures 6A and 6B, regardless of the wheat cultivars.The bulk density level that promoted the highest spike number was 1.12 Mg m -3, Figure 6A.Regarding the spike dry mass, the best bulk density level was 1.16 Mg m -3 , Figure 6B, achieving yield of 4.86 g pot -1 that represents a 60% increase when compared with the maximum bulk density level (1.8 Mg m -3 ).Bonfim-Silva et al. ( 2011) observed a decrease in the structural development and yield of wheat plants when cultivated under the bulk density of 1.30 Mg m -3 , which is a higher value than that found in this study (1.16 Mg m -3 ).This indicates that, in this study, the wheat cultivars were more sensitive to the increasing bulk density.Similar results were obtained by Secco et al. (2009), who verified, under field conditions, a grain yield decrease of wheat and maize plants cultivated in Rhodic Hapludox.
However, Cabral et al. (2012), when studying tropical forages (Brachiaria brizantha 'Piatã' and Panicum maximum 'Mombaça'), observed higher concentrations of nitrogen and phosphorus in leaves from plants cultivated under intermediate soil compaction levels of 1.28 and 1.40 Mg m -3 , respectively.This indicates that a moderate compaction increment often increases soil contact with roots, which contributes to a better nutrient absorption.Nevertheless, according to Bonini et al. (2011), over compaction directly affects crop yield.These authors evaluated the compaction effect on Rhodic Hapludox in Southern Brazil under field conditions, and observed that the compaction level applied by five steamroller passes resulted in a 23% decrease in wheat yield.Regarding the dry mass of aerial part, there was an isolated effect only for bulk density.This variable was adjusted to the regression quadratic model, presenting its maximum yield at 1.05 Mg m -3 Figure 7, regardless of the wheat cultivars.Such low value indicates greater root susceptibility to compaction, as there was lower water and nutrient absorption from that level, resulting in minor production of dry mass of aerial part (Collares et al., 2008).Merotto Jr. and Mundstock (1999) also described a decrease, according to soil compaction, in the wheat dry mass of aerial part.On the other hand, Bonelli et al. (2011), when studying four bulk density levels (1.0; 1.2; 1.4 and 1.6 Mg m -3 ) verified that, for P. maximum 'Mombaça', the dry mass of aerial part presented high susceptibility to increasing bulk density, as also indicated by the resulting linear model found for this variable in this study.
The dry mass of roots of the wheat cultivars BRS Guamirim and IAC 350 was adjusted to the regression linear model; it decreased with increasing bulk density, Figure 8.The comparison between compaction absence (1.0 Mg m -3 ) and highest bulk density level (1.8 Mg m -3 ) showed a 64.19% decrease in its production.These results corroborate that described by Bergamin et al. (2010), who concluded that soil compaction negatively influences the maize root system.We observed that the compacted layer promoted a concentration of roots next to the soil surface due to the imposed restriction to root growth.This result is linked to that previously described for dry mass of aerial part, which decreased from the bulk density of 1.05 Mg m -3 .Regarding the chlorophyll index (evaluated by SPAD readings), there were isolated effects for wheat cultivars and bulk density.The IAC 350 cultivar presented higher chlorophyll index, in comparison with BRS Guamirim (Table 2), regardless of bulk density.
The chlorophyll index response to the bulk density levels was adjusted to a linear regression model, with a value of 50.9 obtained from the lower compaction level (1.0 Mg m -3 ), Figure 9.Such value was similar to that observed by Espindula et al. (2009) for wheat plants as maximum leaf chlorophyll index corresponded to the SPAD reading of 50.
The chlorophyll index decrease with increasing bulk density may have been promoted by lower nitrogen absorption by plants, since the dry mass of roots also decreased as bulk density levels increased, Figure 8.
The compaction process increases bulk density and reduce soil total porosity that will directly affect water dynamics in the soil, thus, nitrogen absorption occurs mainly, by mass flow.The reduction in nitrogen absorption under high bulk density levels was also observed by Cabral et al. (2012).According to Teixeira et al. (2010), wheat yield is positively correlated with leaf chlorophyll index; therefore, our results indicate that high bulk density levels may compromise the wheat grain production in the Cerrado of Mato Grosso State.

Conclusions
High bulk density reduced the production characteristics and chlorophyll index of the wheat cultivars BRS Guamirim and IAC 350 cultivated in the Cerrado Rhodic Hapludox; it may be considered as an evaluation parameter of soil physical quality in production systems.The IAC 350 cultivar presented the best spike production and chlorophyll index, regardless of the bulk density levels, so it is considered a promising cultivar for cultivation in Rhodic Hapludox, Central Brazil.
The highest chlorophyll index observed for the wheat  cultivars BRS Guamirim and IAC 350 corresponded to the bulk density of 1.0 Mg m -3 .

Figure 1 .
Figure 1.Representation of the experiment geographic location in the Brazilian Cerrado, which geographic coordinates are 16º27'49" S and 50º34'47" W, in Rondonópolis, Mato Grosso State, Brazil (A); detail of the greenhouse (B); and experiment overview (C).

Figure 2 .
Figure 2. Experimental unit made of poly (vinyl chloride) and respective dimensions: height (A) and diameter (B).

Figure 2 .
Figure 2. Experimental unit made of poly (vinyl chloride) and respective dimensions: height (A) and diameter (B).

Figure 3 .
Figure 3. Graphical representation of the experimental unit, showing the position of the 0.05 m compacted layer.

Figure 3 .
Figure 3. Graphical representation of the experimental unit, showing the position of the 0.05 m compacted layer.

Figure 4 .
Figure 4. Procedure of soil compaction with the aim of a hydraulic press (A); Detail of the wooden mold used during the compaction process (B); Compacted layer of 0.05 m thickness (C).

Figure 4 .
Figure 4. Procedure of soil compaction with the aim of a hydraulic press (A); Detail of the wooden mold used during the compaction process (B); Compacted layer of 0.05 m thickness (C).

Figure 5 .Figure 5 .
Figure 5. Implementation of the experimental unit: Filling the lower cylinder with soil at the density of 1.0 Mg m -3 (A); Joining both cylinders with duct tape, showing the Figure 5. Implementation of the experimental unit: Filling the lower cylinder with soil at the density of 1.0 Mg m-3 (A); Joining both cylinders with duct tape, showing the compacted layer (B); Filling the remaining volume of the upper cylinder above the compacted layer at the density of 1.0 Mg m-3 (C).

Figure 6 -
Figure 6-Number (A) and dry mass (B) of wheat spikes (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Figure 6 .
Figure 6.Number (A) and dry mass (B) of wheat spikes (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Figure 7 -
Figure 7-Dry mass of aerial part of wheat plants (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Figure 7 .
Figure 7. Dry mass of aerial part of wheat plants (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Figure 8 -
Figure 8-Dry mass of roots of wheat plants (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Figure 8 .
Figure 8. Dry mass of roots of wheat plants (cultivars BRS Guamirim and IAC 350) according to bulk density levels.

Table 1 .
Mean values of spike number and dry mass for BRS Guamirim and IAC 350 wheat cultivars, regardless bulk density.

Table 2 .
Mean of SPAD readings (chlorophyll index) for BRS Guamirim and IAC 350 wheat cultivars, regardless bulk density.
a Means followed by different letters differ from each other by the Tukey test at 5% probability level.