Subsoiling Operations Concurrent to the Distribution of Acidity Amendments in the Soil Profile: The Response from Soybeans

Abstract

In the Brazilian Cerrado region, the soils have significant agricultural potential; however, they are naturally acidic soils due to the constitution of the source material and the high weathering process. Liming has been the main tool for correcting acidity and supplying Ca²⁺ and Mg²⁺ to the soil; however, the effect of the corrector is limited to the soil surface. Consequently, the benefits of liming are restricted to the surface layer of the soil and do not solve the subsurface acidity problems. The objective of this work was to evaluate the effect of the subsoiling operation together with the application of a soil acidity corrector on the nutrition and productivity of soybean crops. The experiment was carried out in the experimental area of the Federal University of Jataí in the state of Goias, Brazil. The experimental design was a randomized block design, consisting of six treatments and four replications. The experiment was implemented in 2018, and it utilized the following treatments: T1: control (without subsoiling and liming); T2: the application of superficial limestone, without incorporation at a dose of 1 Mg ha⁻¹; T3: limestone applied through gravity up to 60 centimeters (cm) deep using a subsoiler, with a spacing between the rods of 0.37 meters (m), at a dose of 2.82 Mg ha⁻¹; T4: limestone applied with the same implement used in T3, at the same dose, but with a spacing between the rods of 0.75 m; T5: subsoiling, without the use of liming; and T6: subsoiling and, subsequently, liming on the surface at a dosage of 1 Mg ha⁻¹. The superficial application of the oxide increases the mineral nutrition of plants, the production components, and the productivity of soybeans. There is no agronomic feasibility for using the subsoiler to incorporate limestone.

1. Introduction

In the agricultural year 2021/22, the world’s soybean production was 355.59 million tons [1]. In this same harvest period, the Brazilian production was estimated at 123.82 million tons, with an average productivity of 3026 kg ha⁻¹ [2].

Soybean is the main crop produced in Brazil, and because of its great economic importance, practices that lead to high productivity are constantly being sought [3]. Brazilian farmers, for the most part, are proficient at adopting new technologies, and this factor is the key contributor to the increase in production quantities in the country [4].

Research studies are frequently found in the literature that highlight the adoption of conservation systems, such as the direct planting system (DPS) and crop rotation, as excellent alternatives for agricultural soil management, in addition to providing increases in soybean productivity [5,6,7].

However, in soils under the DPS, where amendments and fertilizers are applied on the surface, there is a concurrent increment in the levels of some elements that can be toxic to plants, such as Fe²⁺, Mn²⁺, and, mainly, Al³⁺ [8]. In these systems, experiments have demonstrated that improvements in the chemical characteristics of the soil are restricted to the superficial layers, creating an unfavorable environment for the deepening of the root system [3,8,9,10,11].

Physical preparation operations to decompress soils can be used opportunistically to incorporate correctives [12], suggesting that a tillage operation could be used to incorporate limestone to the depths where soil pH restriction occurs [13]. This soil preparation used to incorporate limestone would promote favorable conditions for the growth and development of crops, positively contributing to the volume of soil explored by the roots [14], and the infiltration [15], supply, and cycling of nutrients [16].

Farmers and technical consultants have been looking for alternatives to improve the deeper layers to create a high-quality chemical environment that promotes deep-root growth [17].

The different implements available for soil tillage cause changes in the soil chemical, physical, and biological properties [18]. One of the implements used to incorporate limestone is the plow harrow, which normally incorporates the limestone at a maximum depth of 20 cm. The moldboard plow is another implement used to incorporate these materials, reaching depths of up to 30 cm [18,19]. In evaluations of the feasibility of improving subsurface acidity using residual (undissolved) limestone and tests of whether deep plowing and incorporation can correct subsurface soil acidity, it was observed that the use of a rotary hoe to a depth of 0.25 m significantly increased the soil pH. In addition, the deep incorporation of 6 Mg ha⁻¹ of limestone to a depth of 0.45 m with a rotary hoe also increased the pH and decreased toxic aluminum [20]. These implements, however, completely turn over the soil and as a rule are used only when implementing the system.

In systems that have already been implemented, some farmers use scarifiers and subsoilers, which are pieces of equipment that work on the surface and subsurface of the soil to promote the disintegration of the compacted layers [21]. Such equipment does not disturb the soil, and whilst its usage is subject to much discussion in farming areas, it can be a viable alternative for applying limestone to the subsurface with minimal disturbance [22]. Therefore, it can be an advantageous option for promoting the incorporation of limestone to soil layers where it would have limited impact with distribution only on the soil surface [23].

The relative productivity of grains was assessed, determined by the combined analysis of four crops (soybeans, corn, beans, and wheat) over three years of agricultural production, and it was noted that the direct planting system, involving subsoiling and the application of additional lime on the surface and scarification treatment, promoted higher crop yields over the three years of evaluation compared to the direct planting system without mechanical intervention [24].

The assessments of the practice subsoiling at depths of 0.4, 0.60, and 0.80 m concluded that the subsoiling tested in the study presented efficiencies proportional to the profile disturbance gradient and improved the root environment, evaluated using attributes that describe the dynamics and distribution of water in the soil profile [25].

Therefore, it is hypothesized that the incorporation of limestone in depth with the use of the subsoiler will provide a more significant development of the root system along the soil profile, with a consequent increase in the plant’s ability to absorb water and nutrients, resulting in an increase in soybean grain productivity.

The objective of this work was to evaluate the effect of preparation in subsurface layers of the soil, associated with distribution methods for the corrective agent, on the vegetative development, mineral nutrition, and soybean grain productivity.

2. Materials and Methods

2.1. Characterization of the Experimental Area

The experiment was carried out in the experimental area of the school farm of the Federal University of Jataí, Jataí, in the state of Goiás, with geographic coordinates of 17°92′82′′ S, 51·71′88′′ W and an altitude of 685. The soil in the experimental area is classified as dystroferric red latosol, with a very clayey texture [26]. Before the installation of the experiment, the chemical and textural characterization of the soil was carried out (

Table 1. The chemical and textural characterization of the soil before the experimental setup. Jataí, state of Goias, 2019.

Layer	pH CaCl2	Ca	Mg	Al	H+Al	K	* P	S
cm		---------------- cmolc dm−3 ------------------				--------mg dm−3---------
00–20	4.90	1.87	0.80	0.06	4.80	49.0	3.90	10.90
20–40	4.70	1.00	0.47	0.05	4.10	28.0	1.90	41.10
40–60	5.30	0.76	0.33	0.00	3.10	18.0	1.90	49.50
60–80	5.10	0.81	0.37	0.04	2.90	15.0	1.40	41.50
80–100	5.50	0.71	0.37	0.00	2.60	17.0	1.30	13.20
Layer	O.M.	O.C.	B	Cu	Fe	Mn	Zn	Na
cm	---------- g dm−3 --------		------------------------- mg dm−3 -----------------------------
0–20	29.70	17.2	0.10	24.6	34.9	31.8	37.7	3.80
20–40	-	-	-	20.3	43.1	18.9	9.20	2.40
40–60	-	-	-	19.2	48.4	11.9	3.90	2.50
60–80	-	-	-	20.8	54.0	12.0	3.00	2.30
80–100	-	-	-	21.4	52.3	11.0	3.00	2.60
Layer	CEC	Base Sat.		Al Sat.	Clay	Silt		Sand
cm	cmolc dm−3	----------%----------			-------------g kg−1-------------
0–20	7.60	36.8		2.10	640	75		285
20–40	5.60	27.5		3.10	-	-		-
40–60	4.20	27.1		0.00	-	-		-
60–80	4.10	29.8		3.20	-	-		-
80–100	3.70	30.3		0.00	-	-		-

* P extracted through the Mehlich-1 method. cm: centimeters.

) in the 0–20, 20–40, 40–60, 60–80, and 80–100 cm layers, following the methodologies described in [27].

The summary of the methodologies used to determine the chemical attributes of the soil are described below:

pH: measurement of the effective concentration of H⁺ ions in the soil solution, electronically, using a combined electrode, immersed in a soil suspension: 0.01 mol L⁻¹ CaCl₂ solution in a ratio of 1:2.5;

Calcium, Magnesium, and Aluminum: Exchangeable Ca and Mg are extracted by KCl 1 mol L⁻¹ together with exchangeable Al, titrating, in a fraction of the extract, aluminum with NaOH, in the presence of bromothymol blue as an indicator. In another fraction of the extract, calcium and magnesium are titrated by complexometry with EDTA, using Eriochrome black-T as an indicator. In a third aliquot, calcium is determined by complexometry with EDTA and calconcarboxylic acid as an indicator;

Hydrogen + Aluminum: Extraction using calcium acetate based on the buffering properties of the salt, resulting from the presence of acetate anions. With the pH level set at 7.0, it extracts much of the soil’s potential acidity up to that pH value;

Phosphorus, Potassium, Sodium, and Micronutrients: The Mehlich-1 extracting solution consists of a mixture of HCl 0.05 mol L⁻¹ + H₂SO₄ 0.0125 mol L⁻¹. The use of this solution is based on the solubilization of these elements by the effect of the pH, between 2 and 3, with the role of chlorine being to restrict the reabsorption process of the newly extracted phosphates [28]. For the micronutrients, the soil/extract ratio used was 1:5; meanwhile, for the other elements, it was 1:10 [29];

Sulfur: Extraction of sulfate by phosphate ions (500 mg P/L) dissolved in 2.0 mol L⁻¹ acetic acid and subsequent quantification of available S by measurement in a spectrophotometer [30];

Organic Matter: The determination of the amount of organic matter was carried out by oxidizing it to CO₂ by dichromate ions, in a strongly acidic medium.

The region’s climate is of the Aw type, typical of savannas, according to the Köppen classification, and is characterized by having two well-defined seasons, one dry and cold (autumn and winter), and the other rainy and hot (spring and summer) [31].

All the chemical reagents mentioned above were produced by (LABSYNTH Products for Laboratories Ltda., Diadema, Brazil).

The data relating to average temperatures and rainfall, which occurred during the development of the work, were collected from the Automatic Surface Observation Meteorological Station located at the Federal University of Jataí, owned by the National Institute of Meteorology (Figure 1). In the figure, an accumulated rainfall of 850 mm is illustrated, with an average temperature of 26 °C, during the experimental period for the 2019/2020 harvest, and rainfall of 745 mm, with an average temperature of 23 °C, during the experimental period for the 2020/2021 harvest.

2.2. Experimental Design and Treatments

In this work, the response of the soybean crop to the soil amendment treatments that were previously carried out on October 30, 2018, was evaluated. These treatments were set in a randomized block experimental design, with four replications. The treatments corresponded to T1—control (without subsoiling and liming); T2—superficial application of the amendment, at a dose of 1 Mg ha⁻¹ in a direct sowing system, without incorporation; T3—application of limestone using a dual-function implement, which decompacts the soil (subsoiling) and incorporates the limestone through gravity behind its rods up to 60 cm deep, with a spacing between the rods of 0.37 m, at a dosage of 2.82 Mg ha⁻¹ of the corrector; T4—application of limestone with the same implement used in T3, at the same dose (2.82 Mg ha⁻¹); however, a spacing was used between the rods of 0.75 m; T5—subsoiling (equipment only used to decompact the soil), without the use of amendment material; and T6—subsoiling, followed by surface liming at a dose of 1 Mg ha⁻¹ of amendment material.

For treatment T4, it was taken into account that the implement rod uniformly incorporated the limestone to a depth of 60 cm from the soil surface, with a spacing between the rods of 0.75 m, so that the calculation of the liming for this situation was 2.82 Mg ha⁻¹, adding the values for correction in the layers of 0–20, 20–40, and 40–60 cm according to the requirements presented in the soil analysis (Table 1). Treatment T3 was based on the same principle as that of T4; however, the spacing between the rods was reduced from 0.75 m to 0.37 m.

The calculations to determine the dosages of lime used in each treatment were carried out based on the equation proposed in [32], which establishes that the need for lime, in Mg ha⁻¹, to be applied to the soil, every 20 cm deep, can be obtained via the following equation: NL (0–20 cm) = T × (V2 − V1)/PRNT, where NL represents the need for liming in Mg ha⁻¹; T refers to the cation exchange capacity at a pH = 7.0; V2 represents the required base saturation; V1 is the current base saturation; and PRNT is the actual total neutralization capacity of the limestone used. In this way, for each treatment, the following quantities of limestone were applied to the soil according to the information obtained in the analysis presented in Table 1:

T2—superficial application of limestone: NL (0–20 cm) = 7.60 × (60–36.8)/175 = 1.00 Mg ha⁻¹;

T3—application of limestone using the subsoiler, with a spacing between the rods of 0.37 m: NL (0–20 cm) = 7.60 × (60–36.8)/175 = 1.00 Mg ha⁻¹, added to the recommendation for a layer of 20–40 cm, NL (20–40 cm) = 5.60 × (60–27.5)/175 = 1.04 Mg ha⁻¹; added to the recommendation for a layer of 40–60 cm, NL (40–60 cm) = 4.20 × (60–27.1)/175 = 0.789 Mg ha⁻¹, totaling 2.82 Mg ha⁻¹ of limestone;

T4—application of limestone using the subsoiler, with a spacing between the rods of 0.75 m: NL (0–20 cm) = 7.60 × (60–36.8)/175 = 1.00 Mg ha⁻¹, added to the recommendation for a layer of 20–40 cm, NL (20–40 cm) = 5.60 × (60–27.5)/175 = 1.04 Mg ha⁻¹; added to the recommendation for a layer of 40–60 cm, NL (40–60 cm) = 4.20 × (60–27.1)/175 = 0.789 Mg ha⁻¹, totaling 2.82 Mg ha⁻¹ of limestone;

T6—subsoiling, followed by surface liming: NL (0–20 cm) = 7.60 × (60–36.8)/175 = 1.00 Mg ha⁻¹.

The equipment used in the experiment (Figure 2) has a deposit box, where the amendment material can be placed, which is distributed by gravity over a conveyor belt and then deposited behind the rods, which act to mechanically decompact the soil. The equipment also has a cutting disc to chop vegetable remains on the ground, reducing soil disturbance in direct seeding systems. The experimental objective was to carry out decompaction associated with the correction of the soil profile in strips.

The liming recommendations were carried out to increase the soil base saturation to 60%, following the recommendation for the Cerrado region [33]. As an amendment, calcium and magnesium oxide were used, containing 42% CaO and 18% MgO, and with a PRNT of 175% (Lhoist of Brazil Ltda., Belo Horizonte, Brazil) Each experimental unit was 6 m wide and 8 m long, totaling 48 m². Only the six central lines were considered the useful area, excluding 2 m from each end.

2.3. Field Study Setup and Conduction

2.3.1. Soybean: 2019/2020 Agricultural Year Summer Harvest

The soybean crop was sown on 22 October 2019, with the aid of a Tatu Marchesan brand direct planting seeder (model PST Plus) containing five rows (Tatu Marchesan Ltda., Matão, São Paulo). The DM 68i69 IPRO^® cultivar was sown (Sementes Goiás Ltda., Rio Verde, Brazil) distributing 15 seeds per meter, aiming to obtain a final population of 350,000 plants ha⁻¹. Before sowing, the seeds were treated with Standak Top^® (Basf S.A., São Paulo, Brazil) at a dose of 200 mL of the commercial product (c.p.) for 100 kg of seeds. At the fertilization stage, 40 kg ha⁻¹ of N, 100 kg ha⁻¹ of P₂O₅, and 90 kg ha⁻¹ of K₂O were applied (using formula 02-20-18). The fertilizer used in this experiment was split into two applications of 250 kg ha⁻¹, the first carried out on the day of sowing, and the top dressing carried out 25 days after sowing, with both carried out in broadcast. The fertilization in the soybean cultivation was determined based on the results of the soil analysis following the recommendations described in [33].

The weeds were controlled in post-emergence using the herbicides Roundup^® (Bayer S.A., São Paulo, Brazil) at a dose of 3.5 L ha⁻¹ and Cletodim Nortox^® (Nortox SA., Arapongas, Brazil) at a dose of 0.5 L ha⁻¹ of the commercial products. For disease management, three preventative applications with fungicides were performed. The first spraying was carried out at the R1 reproductive stage, the second application was carried out at the R3 reproductive stage, and the third was carried out at the R6 reproductive stage, using the products Approach^® (0.3 L ha⁻¹) (CORTEVA Agriscience do Brazil LTDA., Barueri, Brazil), Fusion^® (AGROSPEC S.A., Santiago, Chile) (0.58 L ha⁻¹), and Approach^® (0.3 L ha⁻¹), respectively. For pest management, the following products were used: Game^® (UPL do Brazil S.A., Campinas, Brazil) at a dose of 0.15 L ha⁻¹ (applied together with glyphosate and clethodim at the V3 vegetative stage); Mustang^® (FMC Química do Brazil Ltda., Campinas, Brazil) (0.2 L ha⁻¹ applied at the R1 stage); Galil^® (ADAMA Brazil S.A., Londrina, Brazil) (0.4 L ha⁻¹ applied at the R3 stage); and Sperto^® (UPL do Brazil S.A., Campinas, Brazil) (0.3 L ha⁻¹ applied at the R6 stage).

At the R1 reproductive stage, to estimate the relative chlorophyll content in the plant leaf, the chlorophyllometer device model FALKER^® chlorofiLOG CFL1030 (FALKER Automação Agrícola Ltda., Porto Alegre, Brazil) was used; the analysis was carried out on the third trefoil counted from the apex to the base in 10 plants within the useful area of each plot. The experimental unit was 6 m wide and 8 m long, totaling 48 m². Only the six central lines were considered useful areas, excluding 2 m from each end.

For the nutritional analysis of the soybean crop, the third trefoils without petioles counted from the apex to the base of 10 plants within the useful area were collected, also at the R1 stage. The leaves remained in an oven at 60 ºC until attaining a constant weight. Then, they were ground in a knife mill and subjected to a chemical analysis of their macro- and micronutrient contents, according to the methodology described in [34]. At the R8 stage, the heights and stem diameters of 10 plants were evaluated within the useful area of each plot. The soybean harvest was carried out manually in February 2020. After harvest, the height of the insertion of the first pod, number of pods per plant, number of grains per pod, and grains per plant were measured from a sample of 10 plants collected within the useful area of each plot.

The productivity was obtained from mechanical threshing and by measuring the mass of grains from all plants harvested in the useful areas of the plots. The mass of one thousand soybean grains was obtained using the methodology described in [35].

2.3.2. Soybean 2020/2021 Agricultural Year Summer Harvest

On October 22, 2020, the soybean crop was sown with the commercial cultivar DM 68i69 IPRO^® (Sementes Goiás Ltda., Rio Verde, Brazil) using a direct planting seeder containing five rows of the Tatu Marchesan brand, model PST Plus (Tatu Marchesan Ltda., Matão, São Paulo), distributing 350,000 seeds ha⁻¹, at the 0.45 m spacing between rows. The seed treatment was carried out with Standak Top (Basf SA., São Paulo, Brazil) at a dose of 200 mL of the commercial product (c.p.) for 100 kg of seeds. Using the recommendation described in [33], broadcast fertilization was carried out with 400 kg ha⁻¹ of the formula 02-20-18 on 24 October 2020. The Advance 2000 AM18 Vortex (Jacto) (Maquinas Agricolas Jacto S.A., Pompéia, Brazil) sprayer was used for the application of the phytosanitary products.

The weeds were controlled in post-emergence using the herbicides Roundup^® at a dose of 3.5 L ha⁻¹ and Cletodim Nortox^® (Nortox SA., Arapongas, Brazil) at a dose of 0.5 L ha⁻¹ of the commercial products. For pest management, the following products were used: Fastac^® (Basf S.A., São Paulo, Brazil) at a dose of 0.12 L ha⁻¹ and Nomolt^® (Basf S.A., São Paulo, Brazil) at a dose of 0.15 L ha⁻¹ (applied together with glyphosate and clethodim at desiccation and in the V3 vegetative stage); Expedition^® (Basf S.A., São Paulo, Brazil) (0.3 L ha⁻¹ applied at the R1 stage); Sperto^® (UPL do Brazil S.A., Campinas, Brazil) (0.3 L ha⁻¹ applied at the R3 stage); and Galil^® (ADAMA Brazil S.A., Londrina, Brazil) (0.4 L ha⁻¹ applied at the R6 stage). Concerning disease management, three preventative applications with fungicides were carried out. The first spraying was carried out at the R1 reproductive stage, the second application was carried out at the R3 reproductive stage, and the third was carried out at the R6 reproductive stage, using the products Fusion^® (AGROSPEC S.A., Santiago, Chile) (0.58 L ha⁻¹), Approach^® (CORTEVA Agriscience do Brazil LTDA., Barueri, Brazil) (0.30 L ha⁻¹), and Helmstar^® (Helm do Brazil Mercantil Ltda., São Paulo, Brazil) (0.5 L ha⁻¹), respectively.

At the R1 reproductive stage, to determine the relative chlorophyll content in the plant leaves, the chlorophyll meter device model FALKER^® chlorofiLOG CFL1030 (FALKER Automação Agrícola Ltda., Porto Alegre, Brazil) was used, analyzing 10 plants within the useful area of each plot. The experimental unit was 6 m wide and 8 m long, totaling 48 m². Only the six central lines were considered useful areas, excluding 2 m from each end.

For the nutritional analysis of the soybeans, the third trefoil was collected, counting from the apex to the base of 10 plants within the useful area, at the R1 reproductive stage. The leaves remained in an oven at 60 °C until attaining a constant weight. Then, the dried leaves were processed in a mill, with the dried and ground material subjected to chemical analysis to determine the levels of macro- and micronutrients according to the methodology described in [34]. Before harvesting, the plant heights and stem diameters were evaluated via the random analysis of 10 plants within the useful area of each plot. The soybean harvest was carried out manually in February 2021. After the harvest, the height of the insertion of the first pod, number of pods per plant, number of grains per pod, and grains per plant were measured from a sample of 10 plants collected within the useful area of each plot.

The productivity was obtained from mechanical threshing and via measurement of the mass of grains from all plants harvested in the useful areas of the plots. The mass of one thousand soybean grains was obtained using the methodology described in [35].

2.4. Data Statistical Analyses

The analyses were performed in three different mixed models: the first model evaluated the effects of the oxide dose and subsoiling, the second model evaluated the effect of the oxide application depth, and the third model evaluated the spacing between the subsoiler rods. Mixed-model analyses used the lmer function from the lme4 package in the R software 4.2.2 [36]. The first mixed model considered the fixed effects of the oxide, subsoiling, and year, as well as all the interactions among these effects. The blocks were considered as a random effect. The least-squares means were obtained with the emmeans function from the emmeans package in R [37], and they were compared using the Tukey test when more than two means had to be compared. The second model considered blocks as a random effect and the depth and year as fixed effects, as well as the interaction between them. The third model used the stem spacing and year as the fixed effects and the interaction between them, and it also considered blocks as a random effect [38].

3. Results and Discussion

The use of subsoiling and lime application methods did not affect the physiological and morphological components of the soybean crop in the 2019/2020 and 2020/2021 agricultural years (

Table 2. The chlorophyll Falker index (CFI), stem diameter (SD), plant height (PH), and first pod insertion height (FP) as a function of subsoiling and lime application methods for summer soybean cultivation. The values represent the means of the two agricultural years, 2019/2020 and 2020/21.

Treatments	CFI	SD	PH	FP
		mm	cm	cm
Subsoiling liming	47.30 a	3.94 a	42.90 a	18.10 a
Subsoiling + superficial liming	46.70 a	3.97 a	43.10 a	17.60 a
Subsoiling	47.50 a	4.23 a	41.20 a	18.10 a
Control	45.80 a	3.99 a	40.50 a	19.00 a
Depth
Superficial	47.00 a	3.96 a	43.00 a	17.90 a
Incorporated at 60 cm	47.00 a	3.94 a	42.30 a	17.50 a
Spacing between rods
Subsoiling 0.75	48.10 a	4.14 a	44.80 a	19.40 a
Subsoiling 0.37	48.50 a	4.22 a	45.30 a	19.70 a

Means followed by the same letter in the column are similar to each other using the Tukey test (p ≤ 0.05).

). No significant effect was found for the application of the oxide on the surface compared to the application of the oxide in the strips along the soil profile. The localized application of the oxide in the soil profile using the subsoiler with spacings between the rods of 0.75 and 0.37 m did not promote an increase in the morphological and physiological components of the soybean crop in the 2019/2020 and 2020/2021 summer seasons.

In comparison to the control, the application of the oxide on the surface and subsoiling plus the application of surface lime increased the contents of N and Ca in soybean leaves in the average of the 2019/2020 and 2020/2021 harvests (

Table 3. Levels of N, P, Ca, Mg, and S in soybean leaves as a function of the subsoiling and limestone application methods. The values are the means of two agricultural years, 2019/2020 and 2020/2021.

Treatments	N	P	K	Ca	Mg	S
	g kg−1
Subsoiling liming	20.80 a	2.92 a	24.80 a	9.08 a	3.66 a	3.08 a
Subsoiling + superficial liming	18.70 a	3.06 a	24.80 a	9.44 a	3.92 a	3.02 a
Subsoiling	16.70 ab	2.60 a	24.00 a	7.92 ab	3.61 a	2.70 a
Control	12.40 b	2.41 a	23.40 a	5.31 b	3.15 a	2.72 a
Depth
Superficial	19.10 a	2.90 a	22.50 a	8.84 a	3.70 a	2.98 a
Incorporated at 60 cm	16.30 b	2.37 b	19.60 a	6.93 b	3.24 b	2.78 a
Spacing
Subsoiling 0.75	17.20 a	2.35 a	21.20 a	6.43 a	3.37 a	2.69 b
Subsoiling 0.37	16.10 a	2.62 a	21.00 a	7.84 a	3.58 a	3.07 a

Means followed by the same letter in the column are similar to each other using the Tukey test (p ≤ 0.05).

).

The superficial application of the oxide increased the levels of N, P, Ca, and Mg concerning the incorporated application of the oxide, in bands, up to 60 cm in the soil profile.

The use of the subsoiler to incorporate limestone into the soil profile with a spacing between the rods of 0.37 m positively influenced the S content compared to the use of the subsoiler to incorporate limestone with a spacing between the rods of 0.75 m.

The ranges considered suitable for soybean cultivation aiming at the high productivity of the nutrients N, P, Ca, Mg, and S are 29.1–56.7; 2.2–3.7; 7.8–18; 1.8–5.0; and 1.3–4.3 g kg⁻¹, respectively [39].

The treatments that provided the highest foliar N contents (subsoiling plus the application of surface lime and the application of surface lime in a direct seeding system) were observed to be below the range considered ideal for soybeans. The experimental area had not received soybeans for at least eight prior harvests. This fact may have reduced the population of N-fixing bacteria in the area; therefore, even the inoculation carried out at planting was insufficient for reestablishing the population, thus reducing its efficiency.

For the other nutrients, no limitations were identified for the adequate development of the crop.

The increase in the absorption efficiency of some nutrients by plants after the application of corrective material, as observed in this study, has also been frequently reported in the literature.

The superficial application of acidity correctors increases the availability of N-NO₃⁻ [40]. In the literature, a positive correlation was identified between the soil pH and NO₃⁻ levels (r = 0.92). The authors of [41] reported that a change in the pH from a level of 4.8 to 6.5 resulted in a fivefold increase in the NO₃⁻ content in the soil, with increased absorption by the plants. After three years of applying the soil corrector (2002, 2004, and 2010), liming increased the supply of organic matter to the soil through the addition of biomass in variable amounts according to the crop, growing season, and soil depth. The application significantly increased the total organic carbon at the depths of 10–20 and 20–40 cm [42].

The application of limestone is one of the best tools for indirectly increasing the availability of P [43]. In acidic soils that have high levels of iron and aluminum oxides, part of the available P is fixed, making it unavailable for the plant [44]. The use of acidity correctors promotes an increase in the hydroxyl groups (OH-) in the soil solution. The OH- groups from limestone dissolution react partly with excess H⁺, increasing the soil pH, and partly with Al³⁺, which precipitates in the form of Al(OH)₃ [45].

In another study, a greater availability of exchangeable Ca and Mg was observed in the soil 18 months after application, which was reflected in a rise in the concentrations of Ca and Mg in the leaves of the soybean plants [46]. Other authors have observed a linear effect of doses of dolomitic limestone on the Ca and Mg concentrations in upland rice leaves [47].

Brazilian soils exhibit poor levels of S, and most of this element is linked to organic compounds [48]. In this respect, the use of calcium and magnesium oxide is an important tool for increasing the organic matter content of the soil due to the increase in the amount of organic residues from above-ground plant residues and root fragments [42].

A larger number of pods per plant was observed in the treatments where subsoiling was carried out plus the application of surface limestone and the application of surface limestone in a direct seeding system (

Table 4. The final population (PFINAL), pods per plant⁻¹ (PP), grains per plant⁻¹ (GP), grains per pod-1 (GPD), and mass of one thousand soybean grains (MTG) as functions of subsoiling and limestone application methods. The values represent the means of the two agricultural years, 2019/2020 and 2020/21.

Treatments	PFINAL	PP	GP	GPD	MTG
					g
Superficial liming	251,543 a	85.50 a	82.30 a	2.34 a	165 a
Subsoiling + superficial liming	242,593 a	88.10 a	81.70 a	2.25 a	158 b
Subsoiling	252,469 a	79.00 ab	79.40 a	2.25 a	154 b
Control	258,642 a	68.10 b	70.40 a	2.20 a	146 c
Depth
Superficial	247,068 a	86.80 a	92.10 a	2.41 a	166 a
Incorporated 60 cm	237,037 a	74.90 b	87.80 a	2.35 a	159 b
Spacing
Subsoiling 0.75	263,881 a	81.90 a	87.20 a	2.32 a	163 a
Subsoiling 0.37	252,770 a	87.40 a	86.50 a	2.36 a	166 a

Means followed by the same letter in the column are similar to each other using the Tukey test (p ≤ 0.05).

). The increases observed in this study were 29 and 20%, compared to the control.

The application of superficial limestone in a direct seeding system was observed to increase the mass of one thousand soybean grains in comparison to the other treatments. The application of surface limestone increased the mass of one thousand grains by 13, 7, and 4% compared with the control, subsoiling, and subsoiling plus the surface application of limestone, respectively.

The number of pods per plant⁻¹ and the mass of one thousand grains were observed as being greater in the plots where superficial oxide was applied in comparison to the application that was incorporated up to 60 cm into the soil profile.

The use of the subsoiler in the limestone-localized application in the soil profile with spacings between stems of 0.37 and 0.75 m did not increase the final population, number of pods per plant⁻¹, number of grains per plant⁻¹, grains per pod^{- 1}, and mass of one thousand grains.

In a paper found in the literature, management strategies, including scarification and subsoiling associated with the use of calcium and magnesium oxide, limestone, and gypsum, did not change the soybean production components in the 2016/17 and 2017/18 agricultural years [49]. In contrast, a study available examining the correction of soil acidity depending on the modes of limestone incorporation concluded that the method of incorporation affects its efficiency [50]. The components that make up the soybean grain yield are the plant density, number of pods per plant, number of grains per pod, and grain weight [51], which are influenced by the genetic constitution of the cultivar, environmental conditions, and genotype and environment interaction [52]. According to a published study, the number of pods per plant is the component that is most influenced by the production environment. The use of subsoiling and the application of limestone should also be highlighted as having a positive influence on the number of pods per plant [53].

The benefits of liming in the production components (pods per plant⁻¹ and mass of one thousand grains) and nutritional parameters are reflected in the increase in the soybean productivity influenced by the use of the subsoiler and methods of limestone application.

The highest soybean productivity was observed in the plots where surface limestone was applied in a direct seeding system (Figure 3A). Subsoiling plus the application of surface limestone and subsoiling were classified as intermediate treatments, and the lowest productivity was observed in the control. The application of surface lime increased the productivity by 27, 16, and 6% compared to the control, subsoiling, and subsoiling plus surface liming, respectively.

The application of superficial oxide increased the soybean productivity by 236 kg ha⁻¹ in comparison to the treatment in which the oxide was applied incorporated in the strips up to 60 cm in the soil profile (Figure 3B).

The use of the subsoiler in the local application of limestone in the soil profile with a spacing between the stems of 0.37 m increased the soybean productivity by 198 kg ha⁻¹ compared to the use of the subsoiler in the localized application of limestone in the soil profile with a spacing between the rods of 0.75 m (Figure 3C). Using the larger spacing between the stems, the dose per rod was observed to be twice that of the smaller spacing. This greater concentration of the corrective material at the 0.75 m spacing may have excessively raised the soil pH at the locality of the stem’s passage; this fact may have restricted the exploration of the roots in these areas, leaving the plants with a smaller volume of soil to be explored.

The application of the amendment to the surface layer of the soil (from 0 to 10 cm) increases the exchangeable bases (increasing the availability of Ca²⁺ and Mg²⁺) and the pH level and reduces the toxic aluminum (Al³⁺) [10,49], resulting in improvements in the chemical conditions, directly affecting plant growth and nutrition and crop development [52].

A research study analyzing the types of applications and correctors on soil acidity in a long-term experiment concluded that, years after its application, the effects of dolomitic limestone, incorporated or on the surface, were similar, maintaining the Al³⁺ saturation in the soil by less than 10% up to 15 cm and by less than 30% up to 40 cm [53].

Other research studies analyzing soil tillage systems and the application of limestone found that liming increased the N content and the mass of one thousand grains, which resulted in greater productivity [54].

4. Conclusions

The superficial application of the oxide increases the mineral nutrition of plants, the production components, and the productivity of soybeans. There is no agronomic feasibility for using the subsoiler to incorporate limestone.