Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
INTRODUCTION
The consequences of management practices on soil physical quality directly affect crop development in the agricultural system. Thus, the knowledge of the impacts caused by cropping systems becomes essential. Some physical and hydraulic soil properties may help to understand how soil classes and management influence soil quality, reflecting the nutrient, water and root dynamics of the soil profile (Demattê et al., 2017).
Tillage systems can directly affect soil attributes, influencing the physical quality and water availability for plants (Cassol et al., 2017). The non-mobilization of soil and the accumulation of organic material cause significant changes in physical and hydraulic soil properties and may affect soil quality (Stone et al., 2006).
Many soil attributes have great utility in crop planning and soil quality assessment. Although some authors say that soil bulk density (ρ S ) is not a useful indicator of soil quality (Logsdon & Karlen, 2004), this attribute may be modified by soil use and management, resulting in changes in soil function (Corstanje et al., 2017; Caviglione, 2018). The alteration of ρ S affects soil porosity and, consequently, the availability of water to plants as well as the productive capacity and soil quality. The assessment of moisture at field capacity (θ FC ) and at permanent wilting point (θ PWP ) and the available water capacity (AWC) of the soil allow quantifying the availability of water for agricultural crops. They can be used to monitor and to estimate soil water balance (Horne & Scotter, 2016) and, consequently, is valuable to assist in the decisions of management strategies. Tillage systems can alter pore distribution by size, also changing the soil water retention curve (SWRC) and its attributes (Reichardt & Timm, 2012; Cassol et al., 2017).
The physical and hydraulic soil properties and the widespread data dissemination are important and critical for orienting soil management and broader soil quality studies. As commonly reported in literature, the use of no-tillage in place of tillage systems can alter pore distribution by size, also changing the soil water retention curve (SWRC). Other soil properties may also be altered as function of soil management but such information is not gathered together and interpreted in light of soil quality. Thus, we aimed to determine and associate physical and hydraulic soil properties under no-tillage systems in the Subtropical region of Brazil, as well as to raise data to be available for future studies.
MATERIALS AND METHODS
The study was carried out in Paraná and São Paulo States, covering part of the subtropical region of Brazil.
The soil samples were collected in experimental plots from the ABC Foundation (Fundação ABC, 2019), located in the cities of Arapoti-PR, Ponta Grossa-PR and Tibagi-PR, where the soils were classified as Oxisols; also at Castro-PR, in an Inceptisol; Itaberá-SP in an Alfisol; and at Socavão-PR in a Histosol (Soil Survey Staff, 2014). The soil classification was obtained from the 1:10,000 soil maps from a survey conducted by the ABC Foundation following the Brazilian Classification System, and the climate types were identified by Alvares et al. (2013) using the Köppen climate classification (Figure 1).
Examined plots have 50 x 100 m and were located on flat to gently undulating relief. No-tillage systems with crop rotation for over 20 years using soybean and maize during the summer, and wheat and black oats in the winter, were present in the sampled areas.
Three disturbed and undisturbed soil samples were collected at five points in each plot from three layers (0.0 - 0.10, 0.10 - 0.25, 0.25 - 0.40 m). In Ponta Grossa, soil sampling was performed in three years: 2007, 2015 and 2017. In the other sites, the samples were obtained in 2015 and 2017. The disturbed samples were collected with a barrel auger, and the undisturbed samples were collected following the methodology described in Methods of Soil Analysis Manual (Embrapa, 2011) by using stainless steel rings of 59 cm3 (5 cm diameter and 3 cm height). Samples were wrapped up and stored to preserve natural moisture until analyses.
The granulometry and soil particle density analyses were performed with disturbed samples (Figure 2). Clay, silt and sand contents (%) were performed with the pipette method, which is based on the settling velocities of soil particles (Embrapa, 2011). The soil particle density
Figura 1 Sitios de muestreo ubicados en: (A) Arapoti-PR (transición climática Cfa/Cfb); (B) Ponta Grossa-PR (clima Cfb); (C) Castro-PR y distrito de Socavão (clima Cfb); (D) Tibagi-PR (clima Cfb); y, (E) Itaberá-SP (clima Cfa).
(kg m-3) was determined using the volumetric flask method, by adding ethanol (92 ° GL) and stirring the sample until the air was expelled from soil pores (Gubiani et al., 2006). The following analysis were performed with the undisturbed samples: soil bulk density (kg m-3) by the ratio of soil dry mass and the ring volume; volumetric moisture at field capacity (θ FC ; m3 m-3); total, macro and microporosity (m3 m-3), being macropores with diameter larger than 0.05 mm and micropores between 0.05 and 0.0002 mm (Klein & Libardi, 2002); and saturated hydraulic conductivity (mm hour-1) (Embrapa, 2011). The total porosity was considered equal to the volumetric soil moisture at saturation (θ SAT ). The moisture at field capacity (θ FC ) was determined with volumetric rings arranged in the tension table at 0.01 MPa. Soil microporosity was considered equal to θ FC value. The macroporosity was obtained by the difference between θ SAT and θ FC (Fabian & Ottoni Filho, 2000).
The parameters θ r , α, m and n of the Van Genuchten (1980) equation were estimated with the SPLINTEX pedotransfer program, version 1.0 (Prevedello, 1999). For each soil layer of the studied sites, the following input data were required for the SPLINTEX program: clay (%), silt (%), fine sand (%) and coarse sand (%), soil particle density (kg m-3), soil bulk density (kg m-3), volumetric soil moisture at saturation (θ S ; %). With the parameterized Van Genuchten (1980) equation, the soil water retention curve (SWRC) was generated using the statistical software R (R Core Team, 2014),
Figura 2 Distribución granulométrica de las muestras de suelo recolectadas en las capas: (A) 0.0 - 0.10 m; (B) 0.10 - 0.25 m; y, (C) 0.25 - 0.40 m, de sistemas de siembra directa sobre rastrojo en las localidades de Arapoti, Castro, Ponta Grossa, Socavão y Tibagi - Paraná e Itaberá - São Paulo, región Subtropical de Brasil.
for the 0.0 - 0.10 m, 0.10 - 0.25 m and 0.25 - 0.40 m soil layers. It was considered as permanent wilting point (θ PMP ) the volumetric moisture obtained at 1.5 MPa pressure, estimated with the Van Genuchten (1980) equation.
The soil available water capacity (AWC) was determined with the expression:
Where: AWC − soil available water capacity (mm); ( FCi − volumetric soil moisture at field capacity in the i-th layer (m3 m−3); ( PMPi − volumetric soil moisture at the permanent wilting point in the i-th layer (m3 m−3); z i − i-th rooted soil layer depth (mm); n − number of layers considered.
Statistical analyses consisted in the determination of trend and data dispersion of the physical and hydraulic soil properties and graph comparisons. The correlation coefficient was also calculated between the absolute values of physical and hydraulic soil properties of the no-tillage systems using the statistical software R (R Core Team, 2014).
RESULTS AND DISCUSSION
The soil bulk density (ρ S ) ranged between 1400 kg m(3 (Arapoti) to less than 1000 kg m(3 (Tibagi) (Figure 3). According to Reichardt & Timm (2012), clay soils generally present lower ρ S than sandy soils, a fact observed in the present study. Arapoti and Ponta Grossa presented higher ρ S than other localities with higher clay content, like Socavão and Tibagi (Figure 2). Overall, the ρ S tended to be higher in the uppermost layers (0.0 - 0.25 m). Vizioli et al. (2021) studying the effects of long-term tillage systems (since 1989) on soil physical quality and crop yield in an Oxisol, located in Ponta Grossa, Parana State, Southern Brazil, observed higher ρ S values in the layer up to 0.15 m for the conventional and no-tillage systems, and lower values for the strategic tillage (no-tillage with periodical soil chiseling to reduce soil compaction) system. The threshold value separating the high and low ρ S is 1240 kg m(3 for clay soils and about 1650 kg m(3 for sandy soils (Pachepsky & Park, 2015). Although high ρ S values in the soil can be related to compaction, the ρ S values verified in Arapoti (Figure 3) were due to the higher sand content (Figure 2), but being below the limit of 1650 kg m(3 for sandy soils, suggested by Pachepsky & Park (2015). This indicates that ρ S data in our study are within a range that does not compromise the soil physical quality.
The soil particle density (ρ P ) is an attribute that does not depend on the management but on the constitution of the parental material. Overall, ρ P is little variable between soil types and can be between 2300 kg m-3 and 2900 kg m-3. The ρ P values observed in the studied soils (Figure 3) were similar to those indicated in the literature for different soil types (Libardi, 2005). The values ranged between 2330 kg m-3 (Socavão) to 2721 kg m-3 (Itaberá). The lowest ρ P values were observed in the Histosol sampled in Socavão (2215 kg m-3). The organic matter was not measured, however, Histosols are described as predominantly organic soils (Soil Survey Staff, 2014), with low ρ P values. In soils presenting significant quartz content in its composition, ρ P approaches to 2650 kg m-3, similar to the particle density of pure quartz (Reichardt & Timm, 2012).
The average total porosity (α) observed in the studied soils ranged between 0.47 m3
Figura 3 Densidad aparente del suelo (A) y densidad de partículas del suelo (B) de las muestras recolectadas en sistemas de siembra directa sobre rastrojo en las localidades de Arapoti, Castro, Ponta Grossa, Socavão y Tibagi - Estado de Paraná e Itaberá - Estado de São Paulo, región Subtropical de Brasil.
m-3 (Arapoti) and 0.63 m3 m-3 (Castro), in agreement with sandy to clayey soils, respectively (Figure 4). Clayey soils usually have α between 0.52 m3 m-3 and 0.61 m3 m-3 (Libardi, 2005). The soils in Arapoti presented average values lower than 0.52 m3 m-3, probably because this was the soil with the highest sand content among those in this study. Loamy soils usually present α ranging between 0.47 m3 m-3 and 0.51 m3 m-3 (Reichardt & Timm, 2012).
Ideally, the macropores should occupy one-third of the total soil porosity, while the other two thirds should represent micropores (Hillel, 1970; Reichardt & Timm, 2012). Nevertheless, it also depends on the soil type (Freire, 2006). Therefore, the average values found in the studied soils indicate adequate conditions for agricultural cultivation (Figure 4). In most cases, the proportion of macropores and micropores were different than 1/3 and 2/3, respectively, for the studied soils, which may be associated with two factors acting together: i) High clay content in the soils, exception made to the samples in Arapoti; and, ii) The compaction that may occur in the upper layers of no-tilled soils (Scanlon et al., 2008), in addition to other factors such as disaggregation of the soil structure and a decline in organic matter (Bot & Benites, 2005). High clay contents cause the formation of large volumes of micropores, decreasing the proportion of macropores (Hillel, 1970). Compaction, in its turn, can lead to a decrease in the macropore volume, which could explain the very attenuated macropore/micropore proportions in the uppermost soil layers (except in the samples in Tibagi). However,
Figura 4 Porosidad total (A), macroporosidad (B) y microporosidad (C) de las muestras recolectadas en sistemas de siembra directa sobre rastrojo en las localidades de Arapoti, Castro, Ponta Grossa, Socavão y Tibagi - Estado de Paraná e Itaberá - Estado de São Paulo, región Subtropical de Brasil.
considering the limits established by Libardi (2005), Freire (2006) and Pachepsky & Park (2015), it can be considered that the sampled soils in this study were not compacted. According to Bognola et al. (2010), soils with a higher volume of micropores tend to exhibit more saturated pores, which favor nutrient movement to supply the plant. In absolute values, the average volume of soil macropores indicated a higher aeration potential than the ideal: 0.1 - 0.16 m3 m-3 (Baver et al., 1972; Kiehl 1979; Unger et al., 1982; Reichardt & Timm, 2012).
The average values of volumetric moisture at field capacity (θ FC ) ranged between 0.29 m3 m-3 (Arapoti) to 0.50 m3 m-3 (Castro) (Figure 5B), while the values of volumetric moisture at permanent wilting point (θ PWP ) were 0.15 m3 m-3 (Arapoti) - 0.36 m3 m-3 (Castro) following the same trend observed for θ FC (Figure 5C).
The lowest mean AWC values (48.5 mm) were observed in soil samples from Tibagi (Figure 6A) due to the high volumetric moisture values at the permanent wilting point (Figure 5), which is related to a high volume of micropores of small diameter, between 0.05 and 0.0002 mm (Gubiani et al., 2006; Klein & Libardi, 2002; Libardi, 2005). The amplitude between volumetric moisture at field capacity and permanent wilting point, as well as the high values of soil bulk density favored the occurrence of high AWC values in the studied soils (Figure 5A). The AWC values in Ponta Grossa were higher than 60 mm. A similar value was reported by Araujo et al. (2009), who estimated an AWC of 68.3 mm for soybean under no-tillage, at the III and IV growth stages.
Figura 5 Humedad volumétrica del suelo en saturación (θsat; A), capacidad de campo (θ FC ; B) y puntos de marchitez permanente (θ PWP ; C) de las muestras recolectadas en sistemas de siembra directa sobre rastrojo en las localidades de Arapoti, Castro, Ponta Grossa, Socavão y Tibagi - Estado de Paraná, e Itaberá - Estado de São Paulo, región Subtropical de Brasil.
Under no-tillage, the uppermost layer is typically compacted, which has become a recurrent problem (Nawaz et al., 2016; Lima et al., 2018). On the other hand, studies have reported that intensive traffic and soil preparation can cause physical degradation of the soil, influencing its structure and negatively affecting the soil physical indicators (Castioni et al., 2018).
Figura 6 Capacidad de agua disponible del suelo (AWC; A) y conductividad hidráulica saturada (K SAT ; B) de muestras recolectadas en sistemas de siembra directa sobre rastrojo en las localidades de Arapoti, Castro, Ponta Grossa, Socavão y Tibagi - Estado de Paraná e Itaberá - Estado de São Paulo, región Subtropical de Brasil.
The AWC of the system in Ponta Grossa estimated in 2007 (Araujo et al., 2009) and about 10 years later revealed a slight increase in the capacity of the soil to store water (Figure 6A). However, this effect was not detected after a four-year study by Jabro et al. (2016) in a sandy soil submitted to zero tillage. The soil from Ponta Grossa is more clayey, a factor that favors the occurrence of significant changes associated to management (Barbosa et al., 2018). Moreover, we are considering a longer period, when compared to the study of Jabro et al. (2016), which reinforces the importance of long term soil monitoring (Cavalcanti et al., 2019).
Soil saturated hydraulic conductivity (K SAT ) is known to be scale-dependent (Pachepsky & Park, 2015), which can explain the high variability observed among the samples (Figure 6B). The highest average values of K SAT observed in the studied soils were close to 400 mm hour-1 in Arapoti and Socavão, while the lowest K SAT values were observed in Castro (23.7 mm hour-1). Castro also presented the lowest standard deviation among the samples. In a soil section, larger pores may not contribute to water infiltration when they are discontinuous and, therefore, macroporosity may often not correlate with water infiltration rate (Bouma, 1982). In Lepsch et al. (2015) classification, when considering the degree of permeability, all K SAT values found in the studied soils belongs to the “moderate” group. The absence of values considered extremely high or low indicated that the adopted management of the studied soils was not negatively affecting soil quality.
With data of granulometric fraction (% of sand, silt and clay) and soil bulk density (ρ S ) and particle density as input (ρ P ), the SPLINTEX pedotransfer program estimates the parameters of the Van Genuchten equation (Figure 7). With the parameters, it was possible to estimate the soil water retention curve (Figure 8) for the studied soils, which is fundamental for understanding the impact of management on plant available water in the subtropical region we studied and also for future studies (Koekkoek and Booltink, 1999).
The soil water retention curve (SWRC) describes the association between
Figura 7 Parámetros de la ecuación de Van Genuchten obtenidos con el programa de pedotransferencia SPLINTEX en las localidades de la región Subtropical de Brasil: (A) Arapoti; (B) Castro; (C) Ponta Grossa; (D) Socavão; (E) Tibagi; e (F) Itaberá.
volumetric content and the energy with which water is retained in soil particles. The SWRC trend reflected the porosity classes of the evaluated soils. The results showed that as the highest is the amount of micropores, the highest is the tension required to cause water loss from a soil volume, as observed in Castro (Figures 4C and 8B). For soils with a predominance of macropores, such as Arapoti (Figure 4B), water loss occurred in smaller potentials and the asymptote was more evident (Figure 8A).
The highest correlation observed in the present study occurred between the θ FC and the volume of soil micropores (Table 1), which was naturally expected, since the microporosity values were considered the same as the θ FC values, in agreement with
Figura 8 Curva de retención de agua del suelo (SWRC) obtenida con la ecuación de Van Genuchten y parámetros estimados con el programa de pedotransferencia SPLINTEX en localidades de la región Subtropical de Brasil: (A) Arapoti; (B) Castro; (C) Ponta Grossa; (D) Socavão; (E) Tibagi; y, (F) Itaberá.
results obtained by Andrade & Stone (2011). The authors also found a high correlation between these two attributes in their study with more than two thousand soil samples. Fabian & Ottoni Filho (2000), using the same relationship, validated an equation to estimate the θ FC from micropores volume of an Ultisol.
Table 1 Correlation coefficient between the absolute values of soil physical and hydraulic properties of samples collected from no-tillage systems in the localities of Arapoti, Castro, Ponta Grossa, Socavão and Tibagi - Paraná State, and Itaberá - São Paulo State, Subtropical region of Brazil.
A narrow correlation was also found between θ PWP and microporosity. The result indicates that most of the total porosity of the studied soils consisted of micropores, being directly related to soil water retention (Reichardt & Timm, 2012). Carter (1988) found that, under low pressure, the macropore water moves faster and under the high stresses the macropores become full of air while the water in the micropores is still trapped. Therefore, under high pressure, soil water retention is controlled by its micropore volume.
The micropores act directly on soil water retention. As the proportion of soil micropores is higher, more difficult is the water movement. For this reason, the tension required for the water movement in this soil condition is higher. Therefore, soils with a predominance of micropores, when subjected to less water stress for plants, present little water movement and the value of θ FC and θ PWP tend to be high. Thus, these attributes tend to have a close positive correlation (Cavenage et al., 1999; Hillel 1970).
The K SAT of the studied soils were closely correlated with macropores. Generally, K SAT correlates well with soil macropore volume, as it is the medium where water moves most easily along the profile (Mesquita & Moraes, 2004). However, as the studied soils showed a reduced relative volume of macropores, K SAT was positively correlated with the sand content and negative with the clay content.
The θ FC and θ PWP showed a close positive correlation, which is naturally expected since they are attributes that depend on the same sources of variation (Libardi, 2005). The clay, silt and sand contents also showed a significant (p < 0.05) correlation. The negative correlation observed for sand was expected, since it is a fraction of the total soil texture, i.e., when the value of one increases the other tend to decrease.
The availability of data related to the soil physical and hydraulic properties is important for researchers and technicians. Reliable data on physical and hydraulic soil properties are fundamental to guide and assist numerous investigations, as well as contributing to decision-making on the best agricultural practices to be carried out to maintain or improve the soil quality and crop production.
CONCLUSIONS
The crop management under no-tillage for more than 20 years in the subtropical region of Brazil did not result in deterioration of soil physical quality, considering the eleven physical and hydraulic soil properties (granulometric distribution; soil bulk density; soil particle density; total, macro and microporosity; volumetric moisture at field capacity, permanent wilting point and saturation; available water capacity; and saturated hydraulic conductivity) and their acceptable limits established in the literature. Even considering a set of contrasting soil types, the average soil bulk density and soil porosity fall within typical limits that do not restrict the development of annual crops. The proportion between micro and macropores does not differentiate much from the 2:1, considered ideal for good soil aeration and water storage. A variable requiring attention is the saturated hydraulic conductivity, whose figures indicated it is at “moderate” levels.
