Agronomic Efficiency of Phosphorus Fertilizers With Associate Technology in Soy Crop in Cerrado Soil

For commercial soybean production, the presence of phosphorus in the soil is essential, not only to increase productivity but also because it affects basic functions in plant metabolism. Phosphate fertilizers have low efficiency in Cerrado soils. For this reason, the use of technologies associated with phosphate fertilizers is important to increase their efficiency in the soil. The experiment was conducted at Fazenda Rio Brilhante in Coromandel, MG. The experiment was laid out in a randomized block design in a 4 × 4 + 1 factorial arrangement. Treatments consisted of four phosphate fertilizer sources (MAPCONVENCIONAL, MAPpol 1, MAPpol 2, and FOM), four P2O5 doses (40, 80, 120, and 160 kg ha), and the control treatment (no phosphorus application). The soybean cultivar RK8115 IPRO with a plant population of 340 thousand plants per hectare was used. The experimental plots consisted of five planting lines with a spacing of 0.5 meters and a length of 7 meters. In the experiment, phosphorus leaf, morphological evaluation of plant height, stem diameter, number of stems, and pods were measured. For plant biomass, the dry matter of leaves, stems, and pods were determined. For yield evaluation, the calculation of productivity and RAE was compared with MAPCONVENCIONAL. For sources, data were subjected to analysis of variance, and means were separated by Tukey’s rate test (p < 0.05). For P2O5 doses, regression model fitting was performed using the ExpDes.pt package of R Studio software. Morphological assessments responded linearly to P doses, except for DPM. Leaves and pods P content and yield were affected by P dosage and P sources, with a quadratic response to P dosage. Maximum yield efficiency was achieved at P2O5 doses of 122.73; 145.07; 112.03; and 94.71 kg ha, with yields of 3818.30; 4064.67; 4089.03; and 3952.63 kg ha for MAPCONVENCIONAL, MAPpol 2, MAPpol 1, and FOM, respectively. However, MAPpol 1 and FOM provided a dose reduction of 26 and 15 compared to the maximum yield of MAPCONVENCIONAL, respectively.


Introduction
In a bulletin published in February 2020, soybean production in Brazil will reach a historical record of 123.25 million tons, with average productivity of 3,349 kg ha -1 , representing increases of 7.1 and 4.4%, respectively, over the previous harvest, both in production and productivity. The area planted also increased by 2.6% in this crop compared to the previous crop, reaching a total of 36.8 million hectares (CONAB, 2020). For the commercial cultivation of soybeans, the presence of phosphorus in the soil is essential, since it performs basic functions in the metabolism of plants, which can lead to a significant increase in productivity (Araujo, 2005).
contain predominantly clay minerals of the 1:1 type, as well as iron oxides (hematite and goethite) and aluminum (gibsite). These factors increase and control the availability and adsorption of P in the soil (Stauffer, 2016), which contributes to low concentration of the nutrient in the soil (Fink et al., 2014) and negatively affects the diffusive P flux (FDP) in the soil (Raghothama & Karthikeyan, 2005). Additionally, Hopkins (2015) says that the recovery capacity of phosphorus applied by fertilizers in tropical soils ranges from 0.1 to 30%, depending on the pH and type of clay in the soil.
Hence, alternatives to increase the efficiency of phosphate fertilization in Cerrado soils include the use of coating polymers or related technologies and the use of organomineral fertilizers. The accompanying technologies may be gradual or controlled release (Machado & Souza, 2012), the use of polymers with high density of negative charges that bind interfering divalent or trivalent cations (Dunn & Stevens, 2008), or the use of organic acids (Sas et al., 2001), which are components of organophosphate fertilizers. These technologies aim to mitigate the impairment of soil P availability by cations and clays in order to increase the available P in the soil solution for subsequent uptake by plants.
Therefore, the objective of the present work is to evaluate the response of soybean cultivation and the agronomic efficiency of phosphate fertilizers with appropriate technology in the Cerrado, crop 2019/2020, in the Alto Paranaíba region, MG.

Methodology
The experiment was conducted at Fazenda Rio Brilhante in the municipality of Coromandel, MG, at geographic coordinates 18°35′48.5″S 46°53′55.6″W. The climate of the region is classified as Aw with two well-defined seasons, rainy summers and dry winters (Köppen, 1931). The experiment was laid out in a randomized block design (DBC) and in a 4 × 4 + 1 factorial scheme with four replications. Four phosphate fertilizer sources (MAP CONVENCIONAL , MAP pol 1 , MAP pol 2 and organomineral source (FOM-06-26-00)), four P 2 O 5 doses (40, 80, 120 and 160 kg ha -1 ) and one control treatment (no phosphorus application) were used. Soybean cultivar RK8115 IPRO with indeterminate cycle, maturity group 8.1 and plant population of 340 thousand plants ha -1 was used. The experimental plots consisted of five plots, 0.5 m apart and 7 m long, with a total area of 17.5 m 2 .
For the chemical characterization of the soil, a soil sample was collected from the experimental area and then analyzed according to the methods proposed by EMBRAPA (2017).   Note. OBS.: pH in water; Ca 2+ , Mg 2+ and Al 3+ extractor KCl 1 mol L -1 ; P-rem solution CaCl 0.01 mol L -1 containing 60 mg L -1 of P; P-meh and K + mehlich 1 extractor; H+Al extractor Calcium acetate 1 mol L -1 at pH 7.0; M.O. determined by the Walkley-Black method.
To apply the treatments, the planting lines were first marked with a seeder, which served as a reference for opening the furrows. The furrows were opened manually and then the treatments were applied. The furrows were covered, and the sowing was done mechanically with the same machines used for marking the lines. Leaf phosphorus content (P-leaf) was measured and morphological values such as plant height, stem diameter, number of stems and pods, plant biomass (leaves, stems, and pods dry matter-LDM, SDM, and PDM respectively) were determined. In addition, an analysis of productivity and relative agronomic efficiency (RAE) of the treatments was carried out in comparison with MAP CONVENCIONAL .
The P-leaf analysis was conducted when the plants were at the R2 stage (full flowering). Ten clovers per plot were collected for this assessment, with the third fully developed clover leaf counted from the apex of the plant. These were placed in paper bags, properly labeled, and sent to the Laboratory and Central of Soil Fertility Analysis (CeFert), where they were stored in a convection oven at 65 °C until they reached a constant mass. Then the samples were ground in a Willey mill, weighed 0.5 g, and placed in an electric muffle furnace at 500 °C until complete digestion of the organic part of the plant tissue. Then, the extraction method (HNO 3 1 mol L -1 ) and quantification of P content in leaf tissue were carried out according to the vanadate yellow method proposed by EMBRAPA (2009).
Morphological assessments were made when plants were at phenological stage R5, which corresponds to the phenological stage of filling initiation. Two plants per plot were removed and taken to CeFert to evaluate the morphological parameters. Plant height was measured from the crown to the top of the plant using a tape measure labeled in centimeters. The diameter was measured in millimeters between the first and second nodes at the plant neck using a digital caliper. The number of stems and pods was determined by manually counting each plant structure.
LDM, SDM, and PDM assessments were made on the plants used for the morphological assessments. For this purpose, plants were dissected into leaves, stems, and pods, placed in paper bags, and placed in a convection oven at 65 ºC until they reached a constant mass. The samples were then weighed on an analytical balance. For yield evaluation, only the usable area of each plot was recorded, i.e., excluding the end lines and one meter at the beginning and end of each plot, corresponding to a usable area of 7.5 m 2 . After collection, the pods were threshed manually and weighed on an analytical balance and the result was extrapolated to kilo ha -1 .
The relative agronomic efficiency index was calculated according to Prochnow et al. (2003): where, RAE: Relative agronomic efficiency (%); Y Source i : grain productivity with phosphate fertilizer; Y Untreated : grain yield without phosphate fertilizer; Y Standat-source : productivity obtained with the fertilizer MAP CONVENCIONAL .
Data were subjected to the residual normality test (Shapiro-Wilk) followed by analysis of variance. Means were compared using Tukey's test (p < 0.05) for sources, and the regression model was fitted for P 2 O 5 doses using the ExpDes.pt package (Ferreira et al., 2018) in R Studio software (Rstudio Team, 2016).

Results and Discussion
The ratings of LDM, SDM, stem diameter, and plant height ( Figure 2) were affected only by the P 2 O 5 doses, and these responded linearly up to a P 2 O 5 dose of 160 kg ha -1 . The ratings of PDM, P-leaf, number of pods, and productivity were affected by the interaction between the source and dose factors. Note. ¹ Mean scores with same letters in column do not differ by Tukey's test at 0.05 significance.
In the significant interaction between sources and doses for the P-leaf variable (Table 3 and Figure 3), it is noted that when sources were partitioned within doses, there was a significant difference between sources only for the doses of 40 and 160 kg ha -1 P 2 O 5 , and when doses were partitioned within sources, there was a significant quadratic fit for all sources. Accordingly, the highest P-leaf contents were obtained by the application of MAP CONVENCIONAL , FOM, MAP pol 2 and MAP pol 1 at doses of 100.75; 166.00; 106.25 and 181.25 kg ha -1 P 2 O 5 , reaching contents of 3.59; 4.50; 4.07 and 4.38 g kg -1 , respectively.

Conclusion
The use of phosphate sources with associated technologies resulted in an increase in the relative agronomic efficiency and productivity of the soybean crop compared to MAP CONVENCIONAL . Compared to the conventional MAP, the application of FOM showed an increase in agronomic efficiency at 40 and 80 kg ha -1 P 2 O 5 , while with MAP pol1 and MAP pol 2 additives, the highest values of agronomic efficiency were obtained at doses above 80 kg ha -1 P 2 O 5 . At a dose of 120 kg ha -1 P 2 O 5 , the use of MAP pol 1 and MAP pol 2 resulted in an increase in agronomic efficiency of 50 to 40% compared to the conventional MAP, indicating the potential of the additives to increase productivity.