Visual Evaluation of Soil Structural and Sugarcane Root Under Deep Strip-till and Conventional Tillage

The Visual Evaluation of Soil Structure (VESS) is a relatively simple methodology used for comparing management systems and for maintaining or recovering the quality of agricultural soils. The objective of this study was to evaluate the structural soil quality in the production of sugarcane using VESS. Three treatments were established: Deep Strip-till (DST), Conventional Tillage (CT) and Uncultivated area (UC). For DST and CT soil samples were taken from two locations: in-row and inter-row. Soil blocks were extracted from mini-trenches and carefully fragmented into aggregates, whose appearance, resistance, and characteristics of the structural units define quality scores. The density of visible roots was quantified by a grid-based counting method. DST at in-row location had improved the structural quality of the soil, providing greater root growth. Scores of visual soil quality in CT showed no difference between in-row and inter-row locations. Preserved from machinery traffic the in-row trail in CT did not result in benefit to soil quality. Variability in the scores among the replicate blocks for DST in-row suggests that the equipment had produced irregular soil tillage. VESS proved to be a good indicator from which it is feasible to evaluate impacts of agricultural machines and tillage implements on soil quality.


Introduction
Knowledge and quantification of the impacts on soil quality of land use and management are important for the development of sustainable agricultural systems (Doran, 2002). The soil quality can be assessed both for agro-ecosystems aiming the productivity and for natural ecosystems where major aims are maintenance of environmental quality and biodiversity conservation. Thus, measuring the quality of a soil means assigning a value to the soil that expresses its capacity to fulfill a specific function, which in the context of arable agriculture equates to providing a suitable environment for plant development (Bünemann et al., 2018).
Quantification of the quality of natural and degraded soils can be performed by visual methods, in which the soil is described in terms of the size, shape, and porosity of its individual structural units (aggregates). Additional descriptors that correlate with soil quality can be added (Ball et al., 2007). Omuto (2008) found that visual evaluation is an inexpensive and rapid method for identifying the final stages of physical degradation with an accuracy of 60%.
Direct, in-field evaluation of the structural quality of soils from the temperate regions have been made by several methodologies (Ball & Douglas, 2003;Roger-Estrade et al., 2004;Ball et al., 2007;Shepherd, 2009;Mueller et al., 2013;Abdollahi, Hansen, Rickson, & Munkholm, 2015;Leopizzi et al., 2018). However, few studies reporting the evaluation of the quality of from the tropical regions through visual methods have been conducted (Niero et al., 2010;Dechen et al., 2010;Guimarães et al., 2011;Giarola et al., 2013;Moncada et al., 2014;Cherubin et al., 2016). chart. The depictions of soil attributes provided in these charts assist the user in identifying characteristics of the sampled layers of a soil that determine its structural quality. The use of the VESS method has grown because of its simplicity, it requires few equipment for its evaluation, reliability and speed with which the results are obtained (Leopizzi et al., 2018) Incorrect use of agricultural machinery and equipment has been identified as being pointed as cause of the degradation of soil structure (Roque et al., 2010;Soracco et al., 2015) and hence on the root development of crops (Souza et al., 2012). As a result, there has been growing interest in evaluation the quality of soils subjected to different cultivation processes, with a view to defining the technologies most appropriate for specific climate conditions and soils. Effects on the soil properties vary with the type of tillage followed in the management system, and are dependent on the tillage intensity, the traffic of machines, and the type of equipment used (Roque et al., 2010).
The relative ease and speed with which a soil structure evaluation can be performed by visual scoring makes methods like VESS a practical tool for examining how tillage practices affect soil quality (Ball et al., 2007). Thus, the hypotheses of the study are: for crops raised in rows, regardless of the soil tillage method, the soil quality score for samples from within a row, where there is no traffic of agricultural machines, will be better than for inter-row samples. The in-row scores for a soil managed under deep tillage will be better than for the same soil under other systems of tillage. Through their effects on soil structural quality, tillage practices exert an influence on the development of the root system of the crop. This leads to the expectation that soil quality scores will be related to measures of root growth. The objectives of this present study were to compare the structural quality of an Alfisol under deep strip-till and conventional tillage cultivated sugarcane and to search for relationships between soil structural quality and root system development.

Study Area Description of Treatments
The field study was conducted in Piracicaba, State of São Paulo, Brazil (22º41′04″ S and 47º38′52″ W). The climate is subtropical with dry winters, corresponding to classification Cwa within the Köppen scheme (Peel et al., 2007). The mean annual temperature and rainfall are 24 °C and 1273 mm, respectively. The local landscape presents an undulating relief. The soil at the study site was classified as an Alfisol (Soil Survey Staff [USDA], 1999) corresponding to a Nitossolo Vermelho eutrófico latossólico within the Brazilian System for Soil Classification (Santos et al., 2013), with an ochric epipedon, and clay texture (Table 1), with 134.0, 250.5, 115.0 and 500.0 g kg -1 of coarse sand, fine sand, silt and clay content, respectively, in surfaces horizons (A and A/B). The soil is free draining, and contains no stones or rocks. Note. Color = Water state: moist.
The experimental area of 3.50 ha was cultivated with sugarcane under conventional tillage since August 2005. After two cycles, the area was prepared with conventional tillage for the implementation of the experiment. The experiment was carried out in a completely randomized design (CRD), with 3 plots, each one being a treatment. The plots were demarcated within the experimental area, with dimensions of 120 m long by 50 m wide (0.60 ha), one for each treatment. Three replicates were made within each plot. The evaluations were carried out in the cycle of the first ratoon cane.

Soil Tillage Treatments
Treatment 1: Deep Strip-Till (DST)-The deep strip-till of the soil was performed using an equipment that comprises components which simultaneously performed surface tillage, deep tillage with a subsoiler, surface clod breaking, straw windrowing, and the in-row application of lime and fertilizers at different depths (0.40 or 0.80 m). The surface soil tillage (0.00 to 0.40 m) was carried out with a rotary hoe with 16 blades on each wheel; the deep tillage reached to a depth of 0.80 m, and formed the rows of paired plant beds. Agricultural limestone was applied in-row as part of the deep tillage preparation of the plant beds: 2.00 t ha -1 of lime (PRNT of 80%) at a depth of 0.40 m, and 0.80 t ha -1 at a depth of 0.80 m.
Treatment 2: Conventional Tillage (CT)-The soil tillage was carried on with a disc harrow (20 discs of 0.61 m of diameter) for lime incorporation and a leveling harrow to break up clods. These operations extended to a depth between 0.20 and 0.30 m. Thirty-four days after soil tillage (one day before planting), 2 t ha -1 of lime were applied with surface harrowing. The operations on planting day were the application of 0.80 t ha -1 of agricultural gypsum, incorporated with light harrowing, followed by the formation of the furrows.
Due to a lack of native vegetation near the treatment plots, usually used as a reference in studies on the effects of management systems (Argenton et al., 2005), an Uncultivated area (UC) occupied by bamboo vegetation on the same Alfisol was selected as the reference. There was no history of agricultural operations or traffic of machinery through this area, and the bamboo vegetation had been growing undisturbed for approximately 40 years.

Statisti
The soil st of the root procedures The media blocks) w quartiles, compariso number of 2017).

Results
The visual of the aggr  Lower scores (median Sq = 1.26) were observed in the uncultivated area (UC, Figure 2). The dense root system of the bamboo vegetation, plus the accumulated organic litter on the surface of the Alfisol, probably gave to the UC soil a higher content soil organic matter and a greater biological activity than in the tilled soils. Organic matter and biological activity are factors that increase the binding between mineral particles and consequently produce stable soil aggregates. Soil structural quality scores between 1 and 2 are considered indicators of good soil structure by Ball et al. (2007), were also found by Giarola Tormena et al. (2009); Giarola et al. (2010) and Eurich et al. (2014) for areas of native forest which had never experienced any kind of soil tillage and management.
The DST in-row soil presented the median score (Sq = 2.10). Excavation of the mini-trenches and the extraction of the blocks were easier for the DST in-row locations than for DST inter-row, CT in-row, and CT inter-row, and this observation was consistent with the higher median score of the DST in-row soil.
DST in-row aggregates revealed a soil of greater porosity, and high friability. The presence of roots between small and rounded aggregates, which could easily be broken with finger pressure, also contributed to the good quality score for the DST in-row soil. The attributes of the aggregates in the DST in-row soil: small, friable, porous, sub-angular and rounded, have been identified by Shepherd (2009) as characteristics of soils with good structure.
However, as shown in the box plot of the median quality scores (Figure 2), the variability of the scores was very high for the DST in-row. This occurred because in the layer 2 of some of the replications soil was composed of porous round aggregates with sizes between 0.002 and 0.007 m and an abundance of branched roots, mixed with aggregates of 0.001 m around which the roots observed were flattened and showed horizontal growth.
The median scores for the CT in-row and CT inter-row treatments (Sq = 2.36 and 2.24 respectively) were higher than for the DST in-row. Difficulties were found in the extraction of the CT soil blocks from the field, due to the resistance of the soil to the insertion of the blade of the spade, an indication of the presence of a compacted layer. Tormena et al., (2016) also found difficulties in extracting blocks from soil under corn cultivation with mechanized harvesting, where most of them had well-defined horizontal layers, large angular and subangular shaped aggregates, and were difficult to break, which are all signs of soil compaction and poor structure.
CT in-row and CT inter-row presented large subangular aggregates that were resistant to rupture, with a few flattened and grouped roots. These cubic and angular aggregates have higher tensile strength due to their greater density (Guimarães et al., 2011), and are considered a sign of soil compaction. The similar scores of the two CT sampling positions reflected the soil tillage, which used a harrow and a leveler, implements that produced homogenization of the soil, as reported by Tavares et al. (2017) in sugarcane production under conventional tillage. This means that preserved from machinery the in-row trail in this treatment did not result in benefit to visual soil quality.
The highest median score was obtained for the DST inter-row (Sq = 2.58) mainly because of angular aggregates with sizes ranging from 0.001 to 0.002 m and little porosity, low quantity of roots, and those present displayed little branching and were flattened. The soil is compacted, not only because agricultural traffic was directed along the inter-row zone, but also because there was no tillage in the soil on the 1.50 m wide inter-row zone, so that the soil compaction generated in the previous crop cycle was not alleviated.
Two soil layers with structural differences were easily identified regardless of the treatment. The mean depth ranges for the first layer displayed only small variations among treatments (DST in-row = 0.06 m, DST inter-row = 0.08 m CT in-row = 0.05 m, CT inter-row = 0.07 m and UC = 0.05 m). The second layer correspond the inferior limited of first layer until 0.25 m (depth of study).
The structural quality of the soil was always lower for layer 1 than for layer 2 (Figure 3). This differentiation between the two layers was considered due in part to tillage and the traffic of machinery, and partly to the development of the root systems of the sugarcane plants and the incorporation of plant residues on the surface. For the uncultivated soil (UC), the scores of the layers were, as expected, very similar, but also the soil of the deeper layer presented a higher score. Cherubin et al. (2016) studying different land use and management in Brazil, also identified two layers, and the superficial layer (0.00-0.10 m) always presented better visual quality of the soil structure. As well, Tormena et al. (2016) also noted two layers with visually distinct structural quality in medium texture soil cultivated with corn for 7 years with different management systems.  DST Gabriel, 1978, al layer provi g a weighted a inter-row treat r layers. In th ween 3 and 4 in strict the devel abundance of se in the media has a risk for UC) and posi The variability ocks by differe n growing are ure concluded blocks. Vol. 10,No. 11; inter-row; (c