Chemical Attributes of Soil and Response of Wheat to Serpentinite in Direct Seeding System

The serpentinite is an alternative for the correction of soil acidity and is composed of calcium and magnesium silicate. The objective of this study was to evaluate the residual effect of the serpentinite application on soil chemical attributes and the effects on wheat crop productivity in a no-tillage system. The experimental design was a randomized block design, in a subdivided plot scheme, with four replications. The plots were constituted by serpentinite doses (0, 2, 4, 8 and 16 Mg ha) and in the subplots the soil collection layers (0.0-0.10 and 0.10-0.20 m). The chemical attributes of the soil evaluated at 41 months after the application of serpentinite, presented favorable results of the residual power of this corrective. The main results observed are related to the increase of pH, decrease of aluminum content and potential acidity, and increase of Ca, Mg and Si contents, cation exchange capacity (CTC) and base saturation. The residual of the serpentinite in the soil contributed with an improvement in the chemical attributes of the soil, which favored the increase of the dry mass, number of spikes and yield of the wheat crop.


Introduction
Most of the soils of Brazil present problems of acidity, low availability of nutrients and silicon (Si) for the plants, thus requiring constant corrections and fertilization to raise productive potentials.
The serpentinite is a rock of metamorphic origin, ultrabasic mainly formed by dolomite, calcite and silica, therefore a source rich in magnesium and calcium with contents of up to 42% of MgO, being able to contribute to the balance of Ca/Mg ratio of the soil, besides presenting high amounts of silicon (up to 45% SiO 2 ) among other minerals, contributing to the replacement of these minerals in the soil (Friedman, 2013).
According to Tavares et al. (2010) the serpentinite can be defined with a calcium and magnesium silicate with average SiO 2 and MgO contents of 40.56 and 45.70%, respectively.Teixeira et al. (2010) consider the serpentinite with a silicate rock powder with chemical characteristics necessary to be considered a soil corrective.
The serpentinite is therefore soil corrective, source of nutrients and silicon, beneficial element, mainly for the accumulating cultures of this element, such as the tropical grasses, being able to bring benefits due to the increase of the rigidity of the cellular wall, providing better architecture of the plant, increasing photosynthetic efficiency.
In Brazil, studies have reported that potatoes have increased productivity by supplying Si to plants, such as wheat (Sarto et al., 2015), sugarcane (Alovisi et al., 2018) and rice (Tokura et al., 2007).This increase in productivity may be related to changes in soil chemical attributes.
Considering that the use of serpentinite tends to be the agricultural practice in Brazil, a better understanding of the effects of this corrective in the chemical properties of the soil and in the development of the wheat crop is essential to adopt management strategies to improve agricultural production.
In view of the above, the objective was to evaluate the residual effect of the serpentinite on the chemical characteristics of the soil and the effects on wheat yield in the no-tillage system.

Material and Methods
The experiment was carried out under field conditions in the experimental area of the Federal University of Grande dourados (UFGD) in Dourados-MS, geographical coordinates 54°59′13″ west longitude and 22°14′08″ south latitude, with altitude of 434 m, in a typical Dystrophic Red Latosol, clay texture (Santos et al., 2013).According to Köppen, the region's climate is classified as Am, tropical humid or subhumid (Alvares et al., 2013), with an average annual rainfall of 1,400 mm, and average temperatures range from 18 °C to 25 ºC in the colder and respectively.The experimental area was already cultivated with annual crops (soybean, corn and wheat) for more than 20 years.
Before the sowing of wheat, the soil was sampled at depths of 00-0.10 and 0.10-0.20 m.The soil samples were air-dried, sieved, sieved with a 2 mm aperture mesh, and the chemical analysis was carried out where the pH in water, pH CaCl 2 , calcium, magnesium, exchangeable aluminum, phosphorus extracted by Melich-1 and potassium, according to methodology described by Silva (2009).The values of CTC pH 7.0, sum of bases (S) and saturation by bases (V%) were obtained by calculation.For this determination, the methodology described by Korndörfer, Pereira, and Nolla (2004).
The sowing of wheat was carried out with the use of mechanical seeders on 05/25/2016.The BRS 18-Terena wheat variety was used and 160 kg ha -1 of seeds were seeded, with a row spacing of 0.18 m, targeting a population of 500 thousand ha -1 plants.At sowing, no maintenance fertilization was used because fertility levels were adequate for the wheat crop.
For leaf analysis of wheat, 30 leaf leaves were collected at the beginning of flowering.The leaves were dried in an air circulation oven at 60 ºC until reaching constant weight and later milled in a Willey mill to determine macronutrient and micronutrient concentration (Malavolta, Vitti, & Oliveira, 1997).For the determination of silicon, the methodology described by Korndörfer et al. (2004).
At the end of the crop cycle, the entire aerial part was cut to quantify the dry matter of the aerial part (in grams) of the plants, counting the number of tillers, number of spikes, percentage of fertile tillers, number of ears, number of spikelets, number of grains per spikelet, number of grains per spike, mass of 1000 grains and yield.All variables were performed in 6 rows of 1.0 m length randomly at the time of harvest in each experimental unit.The grains were quantified and the data transformed in kg ha -1 to 13% (wet basis).
Data were submitted to analysis of variance and, when there was a significant effect of the serpentinite doses, the regression studies were applied at 5% level, with the aid of the statistical program Sisvar (Ferreira, 2014).

Soil Chemical Attributes
After 41 months of application of the serpentinite, it was observed that there was a significant effect for all chemical attributes of the soil one for interaction between serpentinite doses and depth of soil collection, others only isolated effects of doses and depths.
For the organic matter and soil phosphorus content, there was only a depth factor effect, with higher values in the superficial layer (0.0-0.10 m), with 30.44 and 24.82 g kg -1 of organic matter and 13.60 and 6.39 mg dm -3 phosphorus in the soil, respectively for the layers of 0.0-0.10 and 0.10-0.20 m.These results can be explained by the higher deposition of organic residues that occur in this soil layer over the years and by the low mobility of phosphorus both horizontally and vertically, especially in clay soils (Marschner, 2002).
There was a significant interaction of the factors serpentinite and depth, for the variables calcium, magnesium, sum of bases, cation exchange capacity (CTC) and silicon, indicating that the effects of serpentinite doses on these variables depends on the soil layer studied.The residual of the serpentinite doses influenced in a positive and linear way the Ca contents in the soil, raising the levels from 27.61 to 35.57mmolc dm -3 in the layer 0.0-0.10m depth, but not altered the contents at a depth of 0.10-0.20 m, with a mean of 27.65 mmolc dm -3 (Figure 1A).This increase was relatively high considering the CaO contents present in the serpentinite (0.66%).It is possible that the increase in this intensity could be associated with the absorption of this nutrient by Urochloa (crop present before soil collection) and, after forage decomposition, calcium has been released at the soil surface, since the production of phytomass of Urochloa was higher in the higher doses of serpentinite.According to Pacheco et al. (2013) calcium may be the third most accumulated element in the Urochloa phytomass losing only to N and K.
Mg levels were also influenced significantly and significantly in the two soil layers evaluated (Figure 1B) with relative increases of 56% and 50% in the layers 0.0-0.1 m and 0.1-0.2m, respectively.This increase was expected due to the serpentinite in its chemical composition, 35.07% of MgO due to the presence of dolomite, proving to be an efficient alternative for poor soils in this element.Ramos et al. (2006) and Moraes et al. (2018) also report that the main justification for magnesium increases was due to the high content of this element in magnesium silicates.The Ca and Mg contents were higher in the 0.0-0.10m layer (Figures 1A and 1B), which are due to the solubilization of the corrective and release of Ca and Mg, as well as the mineralization of the nutrients of the residues Urochloa plants deposited on the soil.
The sum of bases was influenced by the residual of the serpentinite doses, where the data adjusted to the increasing linear model at the two depths evaluated (Figure 1C).At the depth of 0.0-0.10m, an estimated 40% increase in the residual dose of 16 Mg ha -1 is observed (Figure 1C).In the 0.10-0.20 m layer the increase was 14%.The highest SB values were observed in the 0.0-0.10m range (Figure 1C).These results are related to the highest values of Ca, Mg and K found in this layer.
For the cation exchange capacity (CTC layer), the data were fitted to the polynomial model, with a minimum CTC value (95.13 mmolc dm -3 ) obtained with the residual estimate of the application of 16 Mg ha-1 in the layer of 0.10-0.20 m and maximum value of CTC (105.67 mmolc dm -3 ) reached with the residual estimate of the application of 12.94 Mg ha -1 in the layer of 0.0-0.10m (Figure 1D).Although no significant values of CTC were obtained for the dose factor alone, CTC in the soil presented higher values in the superficial layer, decreasing with increasing depth, this behavior is due to the higher Ca, Mg, K and organic matter layer of 0.0-0.10 m.
The soil silicon content presented a linear behavior, with soil Si content of 19.66 mg dm -3 , in the residual dose of 16.0 Mg ha -1 , 25% increase in the 0.0-0 layer, 10 m (Figure 1E).The increase in Si availability in soil with the application of silicates is also reported by Sarto et al. (2014).In the 0.10-0.20 m layer, soil Si contents did not fit any mathematical model, presenting a mean of 16.42 mg dm -3 (Figure 1E).Pereira et al. (2007) point out that Si presents low mobility in the soil, which may explain the higher levels in the layer where the serpentinite was added.
For the variables pH in water, pH CaCl 2 , potassium content, aluminum content, potential acidity and base saturation, there was only an isolated effect of the residual of the serpentinite doses.The values of pH in water (Figure 2A) and pH in CaCl 2 (Figure 2B) increased linearly with increasing serpentinite doses, with consequent reduction of aluminum content (Figure 2) and potential acidity (Figure 2E), which confirms the neutralizing action of the serpentinite.According to Korndorfer and Nolla (2003), hydrolysis of the silicate anion present in the serpentinite occurs the release of hydroxyls (OH -), which reacts and neutralizes the H + in solution, raising the pH and precipitating Al 3+ in the form of Al hydroxide [Al(OH) 3 )], low solubility and inactive in soil solution and therefore non-toxic to plants.The reduction of soil acidity with the use of silicates was also observed by other authors such as Alovisi et al. (2018), and Moraes et al. (2018).
For the potassium content, an adjustment of the data to the quadratic function, with a minimum content of K (3.23 mmolc dm -3 ), achieved with the residual application of the dose of 5.61 Mg ha -1 of serpentinite (Figure 2C).Despite the lower value of K in this dose, soil content is still in the high availability range for plants, according to Sousa and Lobato (2004).