Soil Physical Quality in Sugarcane Field Under Cover Crop and Different Soil Tillage Systems

Currently, the management practices employed in Brazilian sugarcane plantations have contribute to soil physical degradation and, few studies considering the effect of cover crop associated with conservationist soil tillage systems to control or even reverse this process. Therefore, with the aim to assess the impact of cover crop and tillage systems on the least limiting water range (LLWR) and the S index in two soils of different textures used for sugarcane production, a fieldwork was carried out in two sugarcane plantations in the state of São Paulo, Brazil. The experimental design is a split-plot with four repetitions. The main factor consisted of soil cover vegetation: cover crop and fallow, and the second factor, the tillage system: minimum tillage and conventional tillage. The data of this study demonstrated that clayey and medium-textured soil are sensitive to the management systems used. The use of cover crop promoted an increase of LLWR (average incremental rate of 105% for clayey and 100% for medium-textured soil) and S index (average incremental rate of 16% for clayey and 10% for medium-textured soil). The maintenance of soil under fallow represented restrictive conditions for the growth/development of the plants due to the degradation of the soil structure. In addition, conservation management systems, such as minimum tillage, resulted in better soil physical quality when associated with cover crop. Finally, the clayey and medium-textured soil, show good S index during the first cycle of sugarcane cultivation.


Introduction
Sugarcane cultivation was introduced in Brazil since the colonial period (between the XVI and XIX centuries) and presents a highlight position in the national economy due to the planted area, nine million hectares (Conab, 2018), which generate directly and indirectly millions of jobs (Neves & Trombina, 2014) and energy potential (Cortez, 2010).
However, sugarcane monoculture, intensive soil tillage, and machinery in-field traffic lead to accelerated soil physical degradation.Conversely, the use of conservation management systems, such as minimum tillage and no-tillage, can help maintaining soil productivity since crop residues are left on the soil surface under both systems, and tillage is minimal and/or restricted to planting rows.Moreover, both systems emphasize the use of cover crop, which improve soil quality conditions.Some soil physical properties, such as bulk density, porosity, penetration resistance and available water are used to characterize soil physical status and to assess the effectiveness of different management systems.Soil physical properties evaluated individually, however, cannot fully explain changes of soil physical conditions-as it can be obtained by a combination of a certain number of them (Silva & Kay, 1997;Tormena, Silva, & Libardi, 1998;Guedes Filho, Blanco-Canqui, & Silva, 2013).
In an attempt to integrate and simplify monitoring of soil physical quality, various authors have used the least limiting water range (LLWR) as an indicator of soil structural quality (Araújo et al., 2013;Gonçalves et al., 2014;Guimarães Júnnyor et al., 2015) and to evaluate the impact of several soil management practices (Tormena et al., 1998;Tormena et al., 2007).The LLWR is defined as the range of soil water content in which limitations to plant root growth correlated with water potential, aeration and mechanical resistance are minimal (Silva & Kay, 1997).Conceptually, the LLWR comprises three factors-soil aeration, soil water retention and soil penetration resistance-that affect plant growth in a single variable.It is used as an indicator of soil structural quality for crop production and as a parameter to characterize the impact of soil management practices on sustainable productivity of soils (Silva et al., 1994).
The S index, defined as the slope of soil water retention curve at its inflection point, is directly related to microstructural porosity of soil.It expresses direct effects of soil management system on soil compaction, soil organic matter and root growth.Therefore, more structural pores are associated with higher S index, which is a desirable trait of a good soil (Dexter, 2004).
Many studies have demonstrated the efficiency of the LLWR and the S index to assess soil physical quality (Dexter, 2004;Leão et al., 2006;Andrade & Stone, 2009;Cavalieri et al., 2011;Betioli Júnior et al., 2012;Gonçalves et al., 2014;Lima et al., 2012).However, only few studies have evaluated the effect of the use of cover crop associated with conservationist soil tillage systems to optimize LLWR and the S index.
We hypothesized that the adoption of cover crop together with minimum tillage during the sugarcane crop reformation period can improve soil physical indicators compared to conventional tillage system with fallow.Therefore, the objective of the study was to assess the impact of cover crop and tillage systems on the LLWR and the S index in two soils of different textures used for sugarcane production.

Study Area
The experiment was carried out in Iracema sugarcane mill, and Santa Fé sugarcane mill, located in Iracemápolis and Nova Europa in the state of São Paulo, respectively.The sites were chosen because the mills are under different edaphoclimatic conditions (Table 1).The soil of Iracema mill was classified as a Rhodic Hapludox (Soil Survey Staff, 2014), with clayey texture according to the texture classification scheme of the Department of Agriculture of the USA (USDA, 2017).The soil of Santa Fé mill was classified as a Typic Hapludox, whit texture medium-textured, according to the same classification systems.

Materials
For the planting of the cover plant, it was used Crotalaria juncea (sunn hemp), IAC KR1, at the dose of 25 kg ha -1 , seeded in rows, spaced every 0.5 m (Table 1).The fallow area, during the development of the cover crop, was subject to twinning of spontaneous species, which was supported by the "seed bank" present in the area.
During the sugarcane planting in the clayey soil, it was 36, 69 and 69 kg ha -1 of the N-P-K, respectively.In he medium-textured soil, 25, 125 and 115 kg ha -1 of the N-P-K were respectively applied in the planting furrows based on the soil analysis performed prior to the experiment installation.
In both areas, the sunn hemp was desiccated and cutting at 0.05 m height.In the clayey soil area this operation was carried out by a tractor Case, model MXM 4 × 4, 110 kW and a strimmers Agritecha.In the medium-textured soil, a tractor Massey Ferguson, model Advanced 275 4 × 2, 202 kW and a strimmers Agritecha were used.
The same treatments were implanted in both areas.For the clayey soil, the conventional tillage plots were prepared through two light harrows using a 36-disc hydraulic grid from Baldan and tractor Case, model MXM, 147 kW and, furrowing at 0.30 m depth using a tractor Valtra, model BH 180 4 × 4, 134 kW and a furrow Driade of two lines.In the medium-textured soil, the equipment'sharrow Santa Izabel with 44-disc and a tractor Valtra, model BT 210 4 × 4, 154 kW.In the plots of minimum tillage, both clayey and medium-textured soil, only the furrowing at 0.30 m depth occurred in which it was used the same equipment used in conventional tillage in each of the respective sites.
For the planting of sugarcane, the stalks were distributed manually in the planting furrows, cut into smaller pieces and later with the aid of a machine, the planting furrows were covered.In the clayey soil was used cover machine of two lines and a tractor Massey Ferguson, model Advanced 275 4 × 2, 202 kW.In the medium-textured soil, was used cover machine DMB also two lines and a tractor New Holland, model TL85E, 65 kWv to cover the planting furrow.

Experimental Design
The experimental design is a split-plot with randomized blocks in which the main factor corresponded to the soil cover vegetation: cover crop and fallow whereas second factor was tillage system: conventional tillage and minimum tillage.In this way, the following treatments combinations were analyzed in this study: i) cover crop with conventional tillage (CCCT); ii) fallow with conventional tillage (FCT); iii) cover crop with minimum tillage (CCMT); iv) fallow with minimum tillage (FMT).
Overall, the experimental area consisted of 16 experimental plots i.e. 4 treatments × 4 replications.Each plot consisted of 15 sugarcane rows.Rows were 34 m long and 1.5 m between consecutive rows.Both sites received the same treatments whereby plots with conventional tillage were made with double harrowing (0.40 m depth) and furrowing (0.30 m depth), and plots with minimum tillage were only furrowed.Sugarcane planting was carried out manually in both experimental sites.Data regarding sunn hemp planting and desiccation, dry-matter (DM) production, sugarcane planting date, fertilization and sugarcane variety are presented in Table 1.Note.CWa= humid temperate; Aw = megathermal or tropical wet.
The experiments started with mechanical elimination of the ratoons and subsoiling to a depth of 0.40 m.Subsoiling was necessary because a compacted layer was detected in the profile, which could have limited root growth in the subsequent cycle.The experiment on clayey soil started in 2013 and on medium-textured soil in 2014, and both sites were collected soil samples at the beginning of the experiment, to characterize the physical attributes of the area (Table 2).
Table 2. Physical properties of soils collected from clayey soil site and medium-textured soil site before the experiment

Parameters Measured
At 90 days after planting of sugarcane undisturbed soil samples were collected at different depths i.e. 0.00-0.10,0.10-0.20,0.20-0.40m, for the assessment of LLWR and S index.Stainless steel rings were used with average diameter and height of 0.045 and 0.050 m, respectively, making an average volume of 88 cm -3 .These samples were saturated by water through gradual increase of water level until two-thirds of the ring height, and the weight of saturated samples were recorded.
Then, the samples were submitted to the following matric potentials: -0.002, -0.008, -0.010 MPa in the tension table; and -0.033, -0.100, -0.500, -1.0, and -1.5 MPa in Richard's chamber with porous plates (Dane & Hopmans, 2002).When the samples reached hydraulic balance for each tension, their weight was recorded to determine their water content.Subsequently, penetration resistance was determined in the laboratory using a Marconi MA-933 bench top electronic penetrometer (Marconi, Piracicaba, São Paulo, Brazil) with constant speed of 1.0 cm min -1 .This device was equipped with a 200-N load cell; rod with cone (base diameter of 4.0 mm) and half angle of 30º; and receiver and interface coupled with a microcomputer to record the data using the equipment's software.
Finally, the samples were oven-dried at about 105 °C for 24 hours, until constant weight was reached.Soil volumetric water content was determined as the ratio of the volume of water removed from the sample in each tension over soil volume of each sample.Bulk density was determined as the weight of dry soil divided by the total ring volume (Embrapa, 2017).
The soil penetration resistance (PR) curve was fitted using the non-linear model recommended by Busscher (1990), as follows (Equation 1).The volumetric moisture was obtained by Topp and Ferré (2002) method.
where, PR is the soil penetration resistance (MPa); Bd is the soil bulk density (Mg m -3 ); θ represents the volumetric water content (m 3 m -3 ); and a, b and c represent the coefficients of the model.
The functional relation between the matric potential and volumetric water content, considering the soil moisture at field capacity (matric potential of -0.01 MPa) and at permanent wilting point (matric potential of -1.5 MPa), was fitted to the model proposed by Silva et al. (1994), according to Equation 2: where, θ is the volumetric water content (m 3 m -3 ), ψ is the matric potential (MPa); Bd is the soil bulk density (Mg m -3 ); d, e and f represent the coefficients of the model.
θ AP was the difference between the volumetric water content at saturation and the air-filled porosity of 0.10 m 3 m -3 , taken as restrictive to the plant growth using Equation 3: where, θ AP is the volumetric water content (m 3 m -3 ) for an aeration porosity of 0.10 m 3 m -3 ; TP is the total porosity (m 3 m -3 ) as described by Blake and Hartge (1986).
The method described by Silva et al. (1994) and Tormena et al. (1998) was used to determine the LLWR, being considered as the difference between the upper and lower water range limits.The upper limit can be associated to θ FC which is the soil moisture when the air-filled porosity is less than 10%.The lower limit can be associated to the soil water content in which the SR value is greater than 2.5 MPa, or to the θ PWP .The critical bulk density value (Bd c ) when LLWR is zero was defined as the intersection of the equations that determine the upper and lower LLWR values.
The S index was calculated using Equation 4, which was formulated by Dexter (2004) and, which uses as a base the adjustment parameters by Van Genuchten (1980) model.
where, S is the value of the slope of the soil water retention curve at its inflection point; θres is the residual water content (m 3 m -3 ); θsat is the saturated water content (m 3 m -3 ); m and n represent empirical parameters of the equation. jas.ccsenet.

Least L
The  The critical limits of the LLWR for studied soils are shown in Figure 1.The values of critical limits used in this study were as follows: field capacity = -0.010MPa, permanent wilting point = -1.5 MPa, aeration porosity = 0.10 m -3 m -3 and penetration resistance = 2.5 MPa, in which the hatched area corresponds to the LLWR.Each soil showed different range of the LLWR, which was due to different soil texture.Soils with higher clay content are characterized by higher water retention due to the aggregation of clay particles: intra-aggregate pores increase the volume of pore space, mainly pores with reduced diameter (Brady & Weil, 2008).In agreement with this statement, Severiano et al. (2011) reported a strong influence of the soil texture on the LLWR, where the increase of the clay content, further the development of textured pores, which, in turn, influences the soil retention water and soil aeration.
For the clayey soil, the lower limit was determined by θ PR and the upper limit by θ FC , up to Bd = 1.13 Mg m -3 .Thereafter, the upper limit of the LLWR for this soil was determined by θ AP .These results demonstrate that up to 1.13 Mg m -3 the soil structure is appropriate.However, for Bd equal or higher than 1.13 Mg m -3 , problems related to anoxia triggered by degradation of soil structure can be expected.Our results corroborate those found by Lima et al. (2012), who observed θ FC as the upper limit, up to Bd = 1.14 Mg m -3 in Hapludox with pasture.The same authors have reported that up to this Bd value the microstructure was stable and preserved, and the pore space was enough for gas exchange.For higher values, however, a reduction of macropores volume was observed.
It was also observed in Figure 1 that an increase in Bd correlated positively with θ PR and negatively with θ AP in clayey soil, indicating that an increase in Bd decreased the LLWR as a function of both the upper limit and lower limit.In addition, for the medium-textured soil, only θ PR (the lower limit) and θ FC (the upper limit) were the limiting variables responsible for restricting the LLWR.
The role of θ PR as the lower limit of the LLWR can be observed for both studied soils, hence demonstrating its direct effect on soil physical properties.High θ PR restricts the range of water availability for root growth and development.These results are aligned with previous studies that indicated that θ PR is the main variable associated with the reduction of soil physical quality resulting in a reduction of LLWR (Tormena et al., 1998;Betioli Júnior et al., 2012;Araujo et al., 2013).Restriction of roots penetration by compacted soil layers may have severe effects on plant growth if the surface soil dries and water supply to the plants is limited by the inability of the roots to tap reserves of water in the subsoil (Materechera et al., 1992).Otto et al. (2011) analyzed the relationship between the spatial distribution of physical soil properties and sugarcane root density in a mechanically harvested area and observed that sugarcane root growth was not affected below PR values of 0.75 MPa, but was decreased significantly between 0.75 and 2.0 MPa.Yet in accordance with these authors, sugarcane root growth was severely restricted when PR > 2.0 MPa.
The LLWR ranged from 0.00 to 0.124 m 3 m -3 for the clayey soil and from 0.00 to 0.040 m 3 m -3 for the medium-textured soil (Figure 2).This range is in the same order of magnitude of studies performed in tropical conditions in Brazil.For example, Cavalieri et al. (2011) observed LLWR values ranging from 0.082-0.122m 3 m -3 for clayey soil and 0.000-0.094m 3 m -3 for a medium-textured soil.Prado et al. (2017) evaluated the soil structural quality after the use of cover crops under no-tillage during the sugarcane crop reformation period in a clayey soil and observed that the LLWR varied between 0.00-0.09m 3 m -3 and between 0.00-0.10m 3 m -3 in depth of 0.15 m e 0.25 m, respectively.
Soil Bd c was found to be 1.18 and 1.65Mg m -3 for clayey and medium-textured soils, respectively.Values of Bd c higher than these probably present limiting conditions to plant development despite the water content in the soil due to structural restrictions.Guimarães Júnnyor et al. (2015) evaluated physical quality of a oxisol (clay 44%) and found Bd c of 1.36 Mg m -3 .According to Petean et al. (2010), low Bd c increases the possibility of Bd achieving critical values (Bd > Bd c ), suggesting a higher incidence of limiting physical conditions for plant development.High Bd values suggest strongly restrictive conditions, mainly in low soil water levels, because they restrict the deepening of the roots, and in a situation of extreme deficits of water in the soil, result in plant water stress (Calonego, Borghi, & Crusciol, 2011;Souza et al., 2015).
However, in the clayey soil Bd was lower than Bd c for all treatments, thus demonstrating good soil structure (Figure 2).The treatment MTCC stood out because it had lower Bd, so more soil micropores, and consequently wider LLWR.On the other hand, the MTF presented higher Bd and narrower LLWR.These results demonstrate a positive effect of cover crop on the LLWR during sugarcane replanting.Traditionally, no-tillage and minimum tillage systems demonstrate limitations related to superficial soil compaction due to natural arrangement of particles and in-field machinery traffic (Tormena et al., 2007). However, Vischi Filho et al. (2016) reported improvements of the LLWR in sugarcane rows under minimum tillage with crop succession after the third sugarcane cropping cycle.The same authors pointed out that sugarcane straw deposited on the soil after jas.ccsenet.

mechanica limitations
In the med Mg m -3 ).T treatments developme species wh compactio conditions confirms th Note.Mac = Macroporosity; Mic = Microporosity; Bd = Bulk density; PR = Soil penetration resistance according impact penetrometer; θ = Volumetric soil moisture corresponding to soil penetration resistance test.
Figure 1.V m), in the Figure 3.

Table 1 .
Location, edaphoclimatic characteristics of the sites and activities in Iracemápolis (Iracema Mill) and Nova Europa(Santa Fé Mill)