Physical quality of sandy soils under orange orchards in Southern Brazil

ABSTRACT

Sandy soils are characterized by low organic matter content and soil water retention and availability. Conventional tillage has been used for the implementation of orange orchards, but it exposes the soil to erosion and promotes accelerated oxidation of organic matter with negative impacts on the soil’s physical quality. The objective of this study was to evaluate the soil physical quality of sandy soils influenced by two soil tillage practices for planting the orange trees in areas after long-time under pastures. Soil sampling was carried out in three experimental areas (one under Lixisol and two under Ferralsol) where the planting of orange trees had been carried out using two tillage practices: (i) conventional tillage in total area; and (ii) localized conventional tillage in a strip corresponding to the orange tree planting lines. A complete randomized design was used with two treatments (total or strip tillage) and four replications. Disturbed and undisturbed soil samples were taken in 2011 (8 to 18 years after the establishment of the treatments) from the 0.00-0.10 and 0.10-0.20 m layers in transects crossing: (i) under canopy projection below orange trees; (ii) under machine wheel tracks in the interrow of the orange trees; and (iii) under grass groundcover between the wheel tracks in the interrow of the orange trees. The following determinations were made for these samples: texture analysis, total organic carbon, soil bulk density, reference bulk density obtained with saturated soil subjected to a 200 kPa, soil resistance to penetration, soil water content, and water retention curves and the least limiting water range. The results suggest that total tillage for the implantation of orange orchards is unnecessary; however, after a long time of establishing orange orchards occurs soil physical quality discontinuity under wheel tracks compared to the other sampling positions. A positive correlation between organic carbon and soil physical quality was identified under the canopy of trees and grass groundcover in the interrow of the orange trees. For similar sand content, the higher soil organic carbon in the Ferralsol provided better physical quality than in Lixisol.

organic carbon; citrus; LLWR; minimum tillage; soil compaction

INTRODUCTION

Approximately 8 % of Brazilian land area is characterized as sandy soil, corresponding to the sand, loamy-sand or sandy-loam textural classes up to a depth of 0.75 m or more, belonging to the Neossolos Quartzarênicos (Arenosols), Latossolos (Ferralsols) and Argissolos (Lixisols), according to the Brazilian System of Soil Classification (Donagemma et al., 2016). In the Northwest region of the state of Paraná, Southern, there are approximately 30,000 km² of sandy soils, corresponding to 16 % of the State’s area (Fidalski and Helbel Junior, 2020). These soils are formed by rocks belonging to the Caiuá Group, which is formed by the Paraná River, Goioerê and Santo Anastácio geological formations (Etchebehere et al., 2007), and referred to locally as the Arenito Caiuá. The classes of soils representative of Caiuá group are the Ferralsols, Lixisols and Arenosols which are in catenas located between the water divisions up to the drainage networks and are characterized, in the same sequence, by an increase in sand content in the superficial layer (Thomaz and Fidalski, 2020). These soils are predominantly fine sands, providing weak aggregation and low water retention capacity (Fidalski and Helbel Junior, 2020). Under agricultural use, there is an even greater reduction in organic matter levels and the physical quality of these soils due to soil compaction (Fidalski et al., 2010).

The predominant use of these sandy soils in the Northwest of Paraná, Southern Brazil, is in line with their agricultural suitability for beef cattle pasture, despite the low cattle stock (Arantes et al., 2018). However, agricultural use has been intensified by cultivating soybean, sugar cane, cassava (Volsi et al., 2020) and orange (Costa et al., 2020). Orange orchards established on pastures using conventional tillage in the orange tree planting lines, i.e., soil tillage in strip 2 m in width, promoted a reduction in soil disturbance of up to 70 % in the tilled area, constituting a soil conservation practice that provides soil and water conservation as well as adequate crop productivity (Mo et al., 2019; Cerdà et al., 2021). Furthermore, these authors found that the maintenance of grassy vegetation in the interrow of the orange trees reduced erosion, and Fidalski et al. (2010) verified the mitigation of the effects of soil compaction by the traffic of machines used in the management of orchards.

In sandy soils, industry representatives have observed that to produce a ton of concentrated orange juice, a greater amount of orange fruit is required when the orchards are planted in soils with higher coarse sand contents (anonymous). Bruand et al. (2005) argue that, in tropical sandy soils, small changes in the texture composition can lead to significant differences in the physical properties of the soils. However, Hondebrink et al. (2017) demonstrated that, in sandy soils, these effects are associated with the magnitude of variability of the soil composition. The results of the physical evaluations of soils under orange orchards cultivated in Lixisol in Southern Brazil (Fidalski et al., 2010) corroborate these findings since the maintenance of cover plants in the interrow of citrus orchards contributes to the improvement of the soil physical quality. Cover crops are management strategies widely practiced in the interrow of orange orchards (Crézé and Horwath, 2021).

Conservation management systems for improving the physical quality of soil under citrus orchards recommend maintaining the interrow vegetation cover, especially with grasses (Hondebrink et al., 2017). The production of biomass and the architecture of the root system of grasses maintain or increase the level of organic carbon, and, therefore, higher water content and lower soil resistance to penetration for the roots of the rootstocks used in citriculture (Homma et al., 2012; Novara et al., 2019).

In the intensively mechanized Brazilian orchards, machine traffic has an impact on the quality physical of soils, as many machines pass per year for orchard management operations (Lima et al., 2004; Fidalski et al., 2010), resulting in up to 45 interrow passes in the first three years of orange tree cultivation. These last authors found that in the first three years after the orange trees were planted, there was the random of machine traffic because there was more space between the lines of orange trees. From the fourth year, after the formation of the orchards, the traffic of machines occurs in permanent tracks to the projections of the canopy of the orange trees, which is identified in the field by the deformation of the soil surface with the formation of two grooves characteristic of the tire wheels of tractors and sprayers.

Studies carried out in orange orchards on Ferralsol and Lixisol sandy soils have found that soil resistance to penetration (PR) was the physical property that most often reduces the least limiting water range (LLWR), mainly in pastures areas (Flávio Neto et al., 2015; Benevenute et al., 2020). In this context, these Ferralsols and Lixisols are differentiated by the presence of Bw and Bt horizons, in which different proportions of sand contents in the surface layer were verified (Fidalski and Helbel Junior, 2020; Thomaz and Fidalski, 2020). The reference bulk density has been frequently used to evaluate the degree of compaction (Dc) under different soils and management systems (Etana et al., 1999; Reichert et al., 2009).

The objective of this study was to evaluate the soil physical quality of sandy soils influenced by two soil tillage practices for planting the orange trees in areas after a long-time under pastures.

MATERIALS AND METHODS

The soil samples were obtained from experiments dedicated to the study of tillage practices for the establishment of the orange trees in Alto Paraná (23° 5’ S; 52° 26’ W), Paranavaí (23° 6’; 52° 25’ W) and Nova Esperança (23° 6’ S; 52° 25’ W), in the Northwest region of the State of Paraná, Southern Brazil, in soils with sandy texture at 0.20 m depth, which were classified as Argissolo Vermelho distrófico (Lixisol; loamy-sand), Latossolo Vermelho distrófico (Ferralsol; loamy-sand) and Latossolo Vermelho distrófico (Ferralsol; sandy-loam) according to Santos et al. (2013) and IUSS Working Group WRB (2006), respectively. The soils are derived from weathered residues of the Alogrupo Rio Paraná from Arenito Caiuá, of the São Bento series, from the Cretaceous period. The three soils showed a predominance for the fine sand fraction and low organic carbon levels in the 0.00-0.20 m layer (Table 1). The climate in the region is classified as Cfa, with an average annual precipitation of 1300-1600 mm and temperature of 20-22 °C, and is characterized by droughts, with rain concentrated in the spring and summer (Alvares et al., 2013).

Table 1
Texture characterization for three soils irrespective of two tillage practices, layers, and transect sampling positions (n = 108 samples)

Descriptive statistics	Fine sand(1)	Coarse sand(2)	Total sand	Silt	Clay
	----------------------------- g kg-1 -----------------------------
Lixisol loamy-sand
Minimum	640	80	840	10	70
Mean	747	132	879	17	104
Maximum	830	220	910	30	150
Ferralsol loamy-sand
Minimum	720	50	850	10	70
Mean	785	102	887	13	101
Maximum	860	190	920	20	140
Ferralsol sandy-loam
Minimum	630	50	750	10	140
Mean	705	99	804	12	183
Maximum	790	180	850	20	230

• ⁽¹⁾ 0.02-0.2 mm. ⁽²⁾ 0.2-2 mm (Santos et al., 2005).

Before the implantation of the treatments, the experimental areas were cultivated with the pastures formed by forages such as Koroniviagrass [Brachiaria humidicola (Syn. Urochloa humidicola)], Bahiagrass (Paspalum notatum) and Palisadegrass [Urochloa brizantha (Syn. Brachiaria brizantha)], planting in 1993, 1994 and 2003, respectively, on Lixisol, Ferralsol loamy-sand and Ferralsol sandy-loam. The field experiments were established in a complete randomized design. The experimental plots consisted of three rows of five orange trees, respectively, about 7 × 4 m. The soil tillage treatments were preceded by surface liming with dolomitic limestone distributed over the total area in doses that increased the base saturation to 70 %. The treatments consisted of two tillage practices: (i) conventional tillage in total area – total tillage; and (ii) conventional tillage in the orange tree planting lines in strip 2 m wide – strip tillage. The soil tillage consisted of plowing with a disc plow at a depth of 0.20 m, followed by leveling with harrows. After establishing the orchards, the management of weeds in the interrow of the orchards was carried out by mechanized mowing without disrupting the soil throughout the experimental period. Planting orange trees, disease and pest control, mowing, liming, fertilizing, and harvesting were carried out using a tractor, making the furrow with a mass of 3,500 kg.

To determine the Dc, disturbed and undisturbed soil samples were collected from the 0.00-0.10 m and 0.10-0.20 m layers from three positions of transects and triplicates by the experimental plot: (i) under canopy projection in lines of orange trees – canopy projection position; (ii) under machine wheel tracks in the interrow of the orange trees – wheel tracks position; and (iii) under grass groundcover between the wheel tracks in the interrow of the orange trees – grass groundcover position, respectively, 1.5, 2.5 and 3.5 m away from the trunk of the orange trees. The soil samples were collected in the experimental plots from three blocks, providing 36 soil samples in each of the experiments, totaling 108 soil samples in the three experiments. Undisturbed samples were collected using steel cylinders 5 cm in height and diameter. Disturbed soil samples were also collected using a Dutch auger, then sieved in a 2 mm diameter sieve and used to determine the texture measurements using the pipette method (sand, silt and clay) and the organic carbon using the Walkley-Black method. Sand content was divided into two fractions: fine sand (0.02-0.2 mm) and coarse sand (0.2-2 mm). In the laboratory, part of these samples was used to determine the reference bulk density, which consisted of placing soil in steel cylinders (2.5 cm high × 7.2 cm in diameter), followed by their saturation for 24 h. Subsequently, the samples were submitted to compression in an automated consolidometer, according to Silva et al. (2007). Soil samples were maintained under a pressure of 200 kPa for 45 min to obtain soil deformation (Håkansson, 1990). Then, the samples were dried in an oven at 105 °C for 48 h to obtain the dry soil mass which was used to compute the soil bulk density (Bd), taken here as the maximum soil bulk density or the reference bulk density (Dr). The Dc was calculated from the values of Bd and Dr: [Dc = (Bd / Dr) × 100 (%)].

During the first half of 2011, thereabout 8, 17 and 18 years after the establishment of the orange orchards, respectively, in Lixisol, Ferralsol loamy-sand and Ferralsol sandy - loam (Table 1), two samplings of undisturbed soil samples were taken simultaneously from the three experiments: one to determine the LLWR and the other for the Dc. To determine the LLWR, 108 undisturbed soil samples were collected from the three experiments using stainless steel cylinders (5 cm in height and diameter). The samples were taken from the 0.00-0.10 and 0.10-0.20 m layers of the canopy projection, the wheel tracks and grass groundcover sampling positions. These samples were used to determine the water content (θ), Dc and penetration resistance (PR). The samples were saturated and allowed to dry naturally in the laboratory at 25 °C. During drying, for the different θ of the soil samples, individual PR measurements were taken with a bench penetrometer (0.04 m diameter cone, 60° angle and 0.05 m high rod) similar to that described by Tormena et al. (1999). Then, to determine the matric potentials (h), a mini-tensiometer with a porous capsule 0.05 m in diameter was introduced into the holes formed by the penetrometer penetration of the rod. For the soil samples that exceeded the reading capacity of the tensiometer, a Dewpoint Potential Meter, model WP4-T (Decagon Devices, Inc., 2007), was used for h measurements, according to Ojeda et al. (2013). Then, soil samples were placed in an oven at 105 °C for 48 h, followed by the determination of water and soil masses to allow the calculation of Bd and θ.

The results of the determinations made on the undisturbed samples were used to adjust the water retention and soil resistance to penetration curves, which were used to estimate the LLWR. The set of 108 Bd (Mg m^-3) observations, with respective θ (m³ m^-3) at different h (hPa) and PR (MPa), independent of tillage treatment, layer, sampling position and soil type was used to obtain curves for water retention and soil resistance to penetration using models described by Silva et al. (1994). The soil water retention curves were determited by equation 1.

$\mathrm{\theta }=\mathrm{aІ}h{\mathrm{І}}^{\mathrm{b}}$ Eq. 1

It was fitted with the model parameters (a and b) and the Bd was integrated into the model using the parameter a (a₀Bd^a1), with a₀ and a₁ being the coefficients. The soil resistance to penetration curve was fitted using the equation 2.

$\mathrm{PR}=c{\mathrm{\theta }}^{\mathrm{d}}B{d}^e$ Eq. 2

After applying a logarithmic transformation at equation 2, we obtained equation 3.

$\mathrm{InPR}=\mathrm{In}c+d\mathrm{In\theta }+e\mathrm{In}Bd$ Eq. 3

in which c, d and e are the fitted parameters (t-test; p<0.05). The fitting of these curves and the LLWR values was performed using the R soil physics package, according to Lima et al. (2020).

The limits of the physical properties used to estimate the LLWR were: air-filled porosity (θ_AFP = 0.10 m³ m^-3), permanent wilting point (θ_PWP = -15000 hPa) and θ_PR (2 MPa), according to Silva et al. (1994). The θ equivalent to field capacity (θ_FC) was estimated at soil h of -30 hPa, as suggested for Brazilian soils by Turek et al. (2020).

The data of Bd, Dc, LLWR and organic carbon were submitted to analysis of variance by the mathematical model of complete randomized blocks, and the mean values of these variables were compared between two tillage practices and three positions of transects by the Tukey's test (p<0.05) and; Pearson’s correlation coefficient (r) was performed for each of the soils between the organic carbon with Bd, Dc and LLWR (n = 36) by t-test (p<0.05), according to Banzatto and Kronka (2006).

RESULTS

Maximum soil density or reference soil bulk density at 200 kPa did not differ significantly, except Lixisols loamy-sand, which presented lower reference soil density than the Ferralsol sandy-loam at 0.10-0.20 m layer (Table 2). There was greater reference soil density in the wheel tracks position than in the canopy projection position for Ferralsol loamy-sand.

Table 2
Mean values for maximum soil bulk density or the reference bulk density at 200 kPa in three sampling positions and two depth layers of three soils

Soil	Canopy projection	Wheel tracks	Grass groundcover
	----------------------------- Mg m-3 -----------------------------
	0.00-0.10 m layer
Lixisols loamy-sand	1.64 a	1.59 a	1.58 a
Ferralsol loamy-sand	1.76 a	1.77 a	1.78 a
Ferralsol sandy-loam	1.85 a	1.88 a	1.84 a
	0.10-0.20 m layer
Lixisols loamy-sand	1.65 a	1.64 a	1.63 a
Ferralsol loamy-sand	1.79 b	1.87 a	1.84 ab
Ferralsol sandy-loam	1.88 a	1.88 a	1.89 a

• Means followed by the same letters in the lines do not differ by the Tukey test (p<0.05).

Soil bulk density (Bd) significantly influenced the water retention and penetration resistance curves (Table 3). With the increase in Bd between 1.24-1.82 Mg m^-3, there was a reduction in LLWR associated with a reduction in θ_AFP and θ_FC and an increase in θ_PWP and θ_PR (Figure 1). The Bd amplitudes were smaller for Ferralsol loamy-sand, followed by Lixisols loamy-sand than Ferralsol sandy-loam (Tables 4, 5 and 6). However, in the LLWR calculation, it was possible to verify common Bd values between 1.57 and 1.80 Mg dm^-3 for the three sandy soils studied. The average Bd values were not influenced by tillage strategies (total and strip tillages) in Lixisol loamy-sand and Ferralsol loamy-sand (Tables 4 and 5), except in Ferralsol sandy-loam in the sampling position relative to wheel tracks and in the 0.00-0.10 m depth layer (Table 6). In this soil, it was found that total tillage increased Bd compared to strip tillage, respectively, from 1.77 to 1.83 Mg dm^-3. However, this increase in Bd did not influence the LLWR after 18 years of establishing the orange orchard. The effect of total tillage on soil compaction was not confirmed by the Dc and soil organic carbon (Tables 4, 5 and 6).

Table 3
Soil water retention and soil resistance to penetration curves adjusted to three soils at the depth of 0.20 m (Equations 1 and 2)

Equations
Soil water retention curve: h = e0.34-0.56*Bb θ-0.25**
Soil resistance to penetration curve: PR = 0.0127 h-1.26** Bd5.03*

• (* p<0.05) and (** p<0.01) significant by t-test.

Figure 1
Values of water content (θ) as a function of soil bulk density (Bd) to the three soils at 0.20 m layer depth irrespective tillage practices and sampling positions at field capacity (θFC = 30 hPa), wilting point (θPWP = 15000 hPa), airfilled porosity (θAFP = 0.10 m3 m-3) and soil resistance to penetration (θPR = 2 MPa). The LLWR corresponds to the least limiting water range.

Table 4
Mean values for soil bulk density (Bd), degree of compaction (Dc), least limiting water range (LLWR) and organic carbon for total tillage and strip tillage in three sampling positions and two depth layers of Lixisol loamy-sand

Bd		Dc		LLWR		Carbon
Total	Strip	Total	Strip	Total	Strip	Total	Strip
------ Mg m-3 ------		------ % ------		------ m3 m-3 ------		------ g kg-1 ------
Canopy projection position (0.00-0.10 m layer)
1.48 a	1.50 a	92 a	90 a	0.12 a	0.11 a	5.2 a	5.2 a
Canopy projection position (0.10-0.20 m layer)
1.56 a	1.54 a	93 a	95 a	0.12 a	0.10 a	3.7 a	3.7 a
Wheel tracks position (0.00-0.10 m layer)
1.62 a	1.63 a	102 a	103 a	0.11 a	0.11 a	5.0 a	5.6 a
Wheel tracks position (0.10-0.20 m layer)
1.74 a	1.76 a	107 a	107 a	0.06 a	0.05 a	3.4 a	3.5 a
Grass groundcover position (0.00-0.10 m layer)
1.56 a	1.56 a	97 a	100 a	0.08 a	0.10 a	4.3 a	5.5 a
Grass groundcover position (0.10-0.20 m layer)
1.66 a	1.66 a	102 a	101 a	0.05 a	0.06 a	3.1 a	3.3 a

• Means followed by the same letters in the lines for tillage practices do not differ by the Tukey test (p<0.05).

Table 5
Mean values for soil bulk density (Bd), degree of compaction (Dc), least limiting water range (LLWR) and organic carbon for total tillage and strip tillage in three sampling positions and two depth layers of Ferralsol loamy-sand

Bd		Dc		LLWR		Carbon
Total	Strip	Total	Strip	Total	Strip	Total	Strip
------ Mg m-3 ------		------ % ------		------ m3 m-3------		------ g kg-1 ------
Canopy projection position (0.00-0.10 m layer)
1.52 a	1.52 a	88 a	85 a	0.16 a	0.16 a	8.3 a	9.1 a
Canopy projection position (0.10-0.20 m layer)
1.63 a	1.60 a	90 a	90 a	0.12 a	0.13 a	6.2 a	6.5 a
Wheel tracks position (0.00-0.10 m layer))
1.71 a	1.73 a	100 a	95 a	0.08 a	0.07 a	8.9 a	9.1 a
Wheel tracks position (0.10-0.20 m layer)
1.77 a	1.76 a	93 a	96 a	0.04 a	0.06 a	5.9 a	6.5 a
Grass groundcover position (0.00-0.10 m layer)
1.51 a	1.57 a	84 a	88 a	0.12 a	0.15 a	8.9 a	8.9 a
Grass groundcover position (0.10-0.20 m layer)
1.65 a	1.64 a	91 a	88 a	0.12 a	0.11 a	5.9 a	6.5 a

• Means followed by the same letters in the lines for tillage practices do not differ by the Tukey test (p<0.05).

Table 6
Mean values for soil bulk density (Bd), degree of compaction (Dc), least limiting water range (LLWR) and organic carbon for total tillage and strip tillage in three sampling positions and two depth layers of Ferralsol sandy-loam

Bd		Dc		LLWR		Carbon
Total	Strip	Total	Strip	Total	Strip	Total	Strip
------ Mg m-3 ------		-------- % --------		------- m3 m-3 -------		------- g kg-1 -------
Canopy projection position (0.00-0.10 m layer)
1.38 a	1.38 a	75 a	74 a	0.18 a	0.20 a	9.3 a	9.3 a
Canopy projection position (0.10-0.20 m layer)
1.58 a	1.56 a	83 a	84 a	0.17 a	0.14 a	7.2 a	6.9 a
Wheel tracks position (0.00-0.10 m layer)
1.83 a	1.77 a	95 a	96 a	0.05 a	0.03 a	8.5 a	8.8 a
Wheel tracks position (0.10-0.20 m layer)
1.77 b	1.80 a	93 a	96 a	0.07 a	0.04 a	6.4 a	6.4 a
Grass groundcover position (0.00-0.10 m layer)
1.64 a	1.66 a	90 a	90 a	0.11 a	0.12 a	8.3 a	9.1 a
Grass groundcover position (0.10-0.20 m layer)
1.68 a	1.70 a	88 a	91 a	0.10 a	0.09 a	6.7 a	6.1 a

• Means followed by the same letters in the lines for tillage practices do not differ by the Tukey test (p<0.05).

Regardless of the tillage practice, there was wide variability in soil physical quality indicators Bd, Dc and LLWR obtained in the transects, with the exception of the LLWR in the 0.00-0.10 m layer in the Lixisol loamy-sand and organic carbon for the three soils in the 0.00-0.10 and 0.10-0.20 m layers (Figure 2). In the sampling position relative to the wheel tracks position, there was an increase in Bd and Dc and a reduction in LLWR compared to the sampling position relative to the canopy projection position in Ferralsol loamy-sand and Ferralsol sandy-loam (Figure 2). The indicators Bd, Dc and LLWR of the position grass groundcover position showed greater variability compared to the other sampling positions (Figure 2). Regardless of the sampled layer, in the sampling positions relative to the canopy projection and the wheel tracks sampling positions, these physical soil quality indicators are equal to or lower than those under the grass groundcover position.

Figure 2
Mean values for soil bulk density – Bd, degree of compaction – Dc, least limiting water range – LLWR and organic carbon at 0.00-0.10 and 0.10-0.20 m depth layer of Lixisol loamy-sand, Ferralsol loamy-sand and Ferralsol sandy-loam, in three sampling position of transects: canopy projection, wheel tracks and grass groundcover, independently of the two tillage practices. Means followed by equal letters in the columns for soil, do not differ by the Tukey test (p<0.05).

For both soils, the Dc ranged from 75 to 107 %, reaching values higher than 100 % in the Lixisol loamy-sand under the wheel tracks sampling position at 0.00-0.10 m and 0.10-0.20 m and grass groundcover sampling position at 0.10-0.20 m (Figure 2), because of lower maximum soil density or reference soil density in Lixisol than in Fersasoils (Table 2). The Dc under wheel tracks were significantly higher than under canopy projection position. Higher Dc was found at 0.00-0.10 and 0.10-0.20 m layer of the Lixisol loamy-sand and at the 0.00-0.10 m layer of the Ferralsol loamy-sand under the wheel tracks and grass groundcover sampling positions compared to the sampling position corresponding to the canopy projection position.

Soil organic carbon contents of the 0.00-0.20 m depth layer of the three soils were negatively correlated (r>0.50) with Dc and positively with LLWR under the canopy projection and grass groundcover sampling positions (Figure 3).

Figure 3
Soil bulk density – Bd, degree of compaction – Dc and least limiting water range – LLWR correlated with the organic carbon by sampling position in each of the three soils, respectively, in three sampling positions: canopy projection, wheel tracks and grass groundcover. Pearson’s correlation coefficient (r): (**p<0.01), (*p<0.05) and (ns: non-significant) by t-test.

DISCUSSION

Soil organic carbon and moisture (Table 1) are within the variability of soils in the Northwest region of the state of Paraná, Southern Brazil (Fidalski and Helbel Junior, 2020). The organic carbon content remained statistically equal between the three sampling positions for the three soils after 8-18 years after the establishment of the orange orchards, in line with Conant et al. (2017).

The effect of tillage practices on Bd in Ferralsol sandy-loam (Table 6) agrees with the results described by Neves et al. (2010) in similar soil in the same region as this study. The presence of forages in the interrow provided advantages found in conservation systems developed for citrus based on maintaining interrow vegetation cover by grasses (Fidalski et al., 2010; Homma et al., 2012; Hondebrink et al., 2017). In line with Mo et al. (2019), the risk of erosion during the implantation phase of the orange groves would decrease in these sandy soils due to the maintenance of pasture grasses between the orange orchards’ interrow.

The results of this study show that tillage conventional should be avoided throughout the area to establish orange orchards under pastures (Neves et al., 2010), considering the results of soil physical quality and the indifference in orange yield in these three experiments established without or with soil tillage in the inter-rows of orange orchards in pastures from Caiuá Sandstone (Auler and Fidalski, 2013).

These results made it possible to characterize a greater horizontal and vertical variability of Bd with the reduction of the sand content of the Lixisol loamy-sand and Ferralsol loamy-sand in relation to the Ferralsol sandy-loam (Figure 2; Tables 1 and 4). An increase in Bd due to machinery traffic in the interrow of orange orchards in Ferralsol sandy-loam (Fidalski et al., 2007) and Lixisol loamy-sand has been reported by Fidalski et al. (2010).

A higher Dc was found under the wheel tracks than in the canopy projection position to three soils, and Dc reached values above 100 % in the Lixisol loamy-sand under wheel tracks position at 0.00-0.10 and 0.10-0.20 m at grass groundcover sampling position (Figure 2), which are associated with lower organic carbon and higher sand contents in Ferralsol sandy-loam (Tables 1, 4, 5 and 6). The reduction in the Dc with an increase in organic matter and clay content agrees with the results of Marcolin and Klein (2011). Differences in the Dc between soil classes were also reported by Etana et al. (1999). Reichert et al. (2009) also found a Dc greater than 100 % for tropical soils with clay contents similar to the Lixisol, and Suzuki et al. (2007) in Alfisol sandy due to high levels of soil compaction in this soil. In our study, the Ferralsols presented the lowest Dc at 0.20 m depth in the canopy projection position (75-94 %) than in the wheel tracks position (94-107 %), allowing us to characterize an adequate physical environment under the canopy projection than grass groundcover sampling positions.

These physical quality indicators did not depend on the organic carbon contents (r<0.50) under the wheel track position. These results also suggest that the organic carbon content also promoted a decrease in the Dc and increase in the LLWR under the canopy projection position in the 0.00-0.10 and 0.10-0.20 m layers and grass groundcover position in the 0.10-0.20 m layer (Figures 2 and 4). Thus, the physical degradation of the soil under the wheel tracks position due to the greater intensity of machine traffic in soils with more sand content, expressed by the reduction in LLWR (Table 1; Figures 1, 2 and 3), was confirmed.

The physical quality indicators of the three soils were not limiting for orange yield (Auler and Fidalski, 2013). Furthermore, the organic carbon contents did not differ between the two tillage practices, sampling positions and soil layers. However, the results suggest that the physical quality indicators of these three sandy soils were dependent on soil organic carbon, similar to the observations of Di Prima et al. (2018), who suggested that tillage treatments should be avoided in established orange orchards.

Management practices for these soils should establish a diminished traffic intensity or employ strategies to run the machinery with greater load-bearing capacity, especially on soils consisting of clayey-sand texture. Covering the soil between orange rows with grass species such as Signalgrass and Palisadegrass is another option for mitigating the damage to the soil structure by the machinery traffic as they provide biological soil decompaction and consequently increase the availability of water for orange trees (Flávio Neto et al., 2015). Recently, other authors have confirmed the mitigation of soil compaction through management with grasses {Signalgrass [Brachiaria decumbens (Syn. Urochloa decumbens)] and Palisadegrass)} in Brazilian tropical soils (Oliveira et al., 2016; Silva et al., 2021). Thus, strategies to reduce the intensity of soil tillage during the establishment of orchards, maintenance of living and dead vegetation cover in the interrow and better distribution of machinery traffic in orchard interrows (Neves et al., 2010; Azevedo et al., 2020) contribute to physical and water improvements through conventional tillage only in the orange tree planting lines to establishment of orchards in pastures, because of the effects of grasses in maintaining the physical quality of the soil (Hondebrink et al., 2017).

CONCLUSIONS

Physical quality indicators of the sandy soils suggest that soil tillage for the establishment of orange orchards is unnecessary in areas previously under pastures. After establishing orange groves for a long time, machinery traffic increases the Bd, the Dc and reduces the LLWR, regardless of soil and organic carbon content. There was a positive and significative correlation between organic carbon and soil physical quality under the canopy projection position and grass groundcover sampling positions. For similar sand contents, the higher organic carbon content in Ferralsol showed less compaction in the orange orchard than Lixisol.

The conventional tillage in total area compromised the physical quality expressed by the Bd in the Ferralsol sandy-loam under the wheel tracks position. The variability of the indicators Bd, Dc and LLWR measured in the transects showed discontinuity in the transects, with better soil physical quality under the canopy projection position compared to the sampling positions relative to the wheel tracks position. Due to the similarities between the sand and organic carbon contents, Ferralsol presented better physical quality than Lixisol.

REFERENCES

• Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G. Köppen’s climate classification map for Brazil. Meteor Z. 2013;22:711-28. https://doi.org/10.1127/0941-2948/2013/0507
  » https://doi.org/10.1127/0941-2948/2013/0507
• Arantes AE, Couto VRM, Sano EE, Ferreira LG. Livestock intensification potential in Brazil based on agricultural census and satellite data analysis. Pesq Agropec Bras. 2018;53:1053-60. https://doi.org/10.1590/S0100-204X2018000900009
  » https://doi.org/10.1590/S0100-204X2018000900009
• Auler APM, Fidalski J. Uso do solo em sistemas conservacionistas para o cultivo de perenes. In: Anais III Reunião Paranaense de Ciência do Solo; maio 2013; Londrina. Londrina: Instituto Agronômico do Paraná, Sociedade Brasileira de Ciência do Solo, Núcleo Estadual do Paraná; 2013. p. 401-8 [cited 2022 Jan 06]. Available from: https://sbcs-nepar.org.br/wp-content/uploads/2020/02/anais-iii-rpcs.pdf
  » https://sbcs-nepar.org.br/wp-content/uploads/2020/02/anais-iii-rpcs.pdf
• Azevedo FA, Almeida RF, Martinelli R, Próspero AG, Licerre R, Conceição PM, Arantes ACC, Dovis VL, Boaretto RM, Mattos Jr D. No-tillage and high-density planting for tahiti acid lime grafted onto flying dragon trifoliate orange. Front Sustain Food Syst. 2020;4:108. https://doi.org/10.3389/fsufs.2020.00108
  » https://doi.org/10.3389/fsufs.2020.00108
• Banzatto DA, Kronka SN. Experimentação agrícola. 4. ed. Jaboticabal: Funep; 2006.
• Benevenute PAN, Morais EG, Souza AA, Vasques ICF, Cardoso DP, Sales FR, Severiano EC, Homem BGC, Casagrande DR, Silva BM. Penetration resistance: An effective indicator for monitoring soil compaction in pastures. Ecol Indic. 2020;117:106647. https://doi.org/10.1016/j.ecolind.2020.106647
  » https://doi.org/10.1016/j.ecolind.2020.106647
• Bruand A, Hartmann C, Lesturgez G. Physical properties of tropical sandy soils: A large range of behaviours. Khon Kaen: Hal Open Science; 2005 [cited 2022 Jan 6]. Available from: https://hal-insu.archives-ouvertes.fr/hal-00079666/document
  » https://hal-insu.archives-ouvertes.fr/hal-00079666/document
• Cerdà A, Novara A, Moradi E. Long-term non-sustainable soil erosion rates and soil compaction in drip-irrigated citrus plantation in Eastern Iberian Peninsula. Sci Total Environ. 2021;787:147549. https://doi.org/10.1016/j.scitotenv.2021.147549
  » https://doi.org/10.1016/j.scitotenv.2021.147549
• Conant RT, Cerri CEP, Osborne BB, Paustian K. Grassland management impacts on soil carbon stocks: A new synthesis. Ecol Appl. 2017;27:662-8. https://doi.org/10.1002/eap.1473/full
  » https://doi.org/10.1002/eap.1473/full
• Costa GV, Neves CSVJ, Telles TS. Spatial dynamics of orange production in the state of Paraná, Brazil. Rev Bras Frutic. 2020;42:e-525. https://doi.org/10.1590/0100-29452020525
  » https://doi.org/10.1590/0100-29452020525
• Crézé CM, Horwath WR. Cover cropping: a malleable solution for sustainable agriculture? Meta-analysis of ecosystem service frameworks in perennial systems. Agronomy. 2021;11:862. https://doi.org/10.3390/agronomy11050862
  » https://doi.org/10.3390/agronomy11050862
• Decagon Devices, Inc. WP4 Dewpoint PotentiaMeter: Operator’s manual: for models WP4 and WP4-T. Version 5. Pullman: Decagon Devices, Inc.; 2007 [cited 2022 Jan 6]. Available from: http://www.ictinternational.com/content/uploads/2017/04/WP4-Operators-Manual.pdf
  » http://www.ictinternational.com/content/uploads/2017/04/WP4-Operators-Manual.pdf
• Di Prima S, Rodrigo-Comino J, Novara A, Iovino M, Pirastru M, Keesstra S, Cerdà A. Soil physical quality of citrus orchards under tillage, herbicide, and organic managements. Pedosphere. 2018;28:463-77. https://doi.org/10.1016/S1002-0160(18)60025-6
  » https://doi.org/10.1016/S1002-0160(18)60025-6
• Donagemma GK, Freitas PL, Balieiro FC, Fontana A, Spera ST, Lumbreras JF, Viana JHM, Araújo Filho JC, Santos FC, Albuquerque MR, Macedo MCM, Teixeira PC, Amaral AJ, Bortolon E, Bortolon L. Characterization, agricultural potential, and perspectives for the management of light soils in Brazil. Pesq Agropec Bras. 2016;51:1003-20. https://doi.org/10.1590/s0100-204x2016000900001
  » https://doi.org/10.1590/s0100-204x2016000900001
• Etana A, Håkansson I, Zagal E, Bučas S. Effects of tillage depth on organic carbon content and physical properties in five Swedish soils. Soil Till Res. 1999;52:129-39. https://doi.org/10.1016/S0167-1987(99)00062-8
  » https://doi.org/10.1016/S0167-1987(99)00062-8
• Etchebehere MLC, Saad AR, Fulfaro VJ. Análise de bacia aplicada à prospecção de água subterrânea no Planalto Ocidental Paulista, SP. Geociências. 2007;26:229-47.
• Fidalski J, Helbel Junior C. Available water content for the management of irrigated crops in the Northwestern Region of Parana State. Rev Bras Agric Irr. 2020;14:3976-86. https://doi.org/10.7127/rbai.v14n101152
  » https://doi.org/10.7127/rbai.v14n101152
• Fidalski J, Tormena CA, Scapim CA. Espacialização vertical e horizontal dos indicadores de qualidade para um Latossolo Vermelho cultivado com citros. Rev Bras Cienc Solo. 2007;31:9-19. https://doi.org/10.1590/S0100-06832007000100002
  » https://doi.org/10.1590/S0100-06832007000100002
• Fidalski J, Tormena CA, Silva AP. Least limiting water range and physical quality of soil under groundcover management systems in citrus. Sci Agr. 2010;67:448-53. https://doi.org/10.1590/S0103-90162010000400012
  » https://doi.org/10.1590/S0103-90162010000400012
• Flávio Neto J, Severiano EC, Costa KAP, Guimarães Junnyor WS, Gonçalves WG, Andrade R. Biological soil loosening by grasses from genus Brachiaria in croplivestock integration. Acta Sci-Agron. 2015;37:375-83. https://doi.org/10.4025/actasciagron.v37i3.19392
  » https://doi.org/10.4025/actasciagron.v37i3.19392
• Håkansson I. A method for characterizing the state of compactness of the plough layer. Soil Till Res. 1990;16:105-20. https://doi.org/10.1016/0167-1987(90)90024-8
  » https://doi.org/10.1016/0167-1987(90)90024-8
• Homma SK, Tokeshi H, Mendes LW, Tsai SM. Long-term application of biomass and reduced use of chemicals alleviate soil compaction and improve soil quality. Soil Till Res. 2012;120:147-53. https://doi.org/10.1016/j.still.2012.01.001
  » https://doi.org/10.1016/j.still.2012.01.001
• Hondebrink MA, Cammeraat LH, Cerdà A. The impact of agricultural management on selected soil properties in citrus orchards in Eastern Spain: A comparison between conventional and organic citrus orchards with drip and flood irrigation. Sci Total Environ. 2017;581-582:153-60. https://doi.org/10.1016/j.scitotenv.2016.12.087
  » https://doi.org/10.1016/j.scitotenv.2016.12.087
• IUSS Working Group WRB. World reference base for soil resources 2006: A framework for international classification, correlation and communication. Rome: Food and Agriculture Organization of the United Nations; 2006. (World Soil Resources Reports 103). Available from: https://www.fao.org/3/a0510e/a0510e.pdf
  » https://www.fao.org/3/a0510e/a0510e.pdf
• Lima CLR, Silva AP, Imhoff S, Lima HV, Leão TP. Heterogeneidade da compactação de um Latossolo Vermelho-Amarelo sob pomar de laranja. Rev Bras Cienc Solo. 2004;28:409-14. https://doi.org/10.1590/S0100-06832004000300001
  » https://doi.org/10.1590/S0100-06832004000300001
• Lima RP, Tormena CA, Figueiredo GC, Silva AR, Rolim MM. Least limiting water and matric potential ranges of agricultural soils with calculated physical restriction thresholds. Agric Water Manag. 2020;240:106299. https://doi.org/10.1016/j.agwat.2020.106299
  » https://doi.org/10.1016/j.agwat.2020.106299
• Marcolin CD, Klein VA. Determinação da densidade relativa do solo por uma função de pedotransferência para a densidade do solo máxima. Acta Sci Agron. 2011;33:349-54. https://doi.org/10.4025/actasciagron.v33i2.6120
  » https://doi.org/10.4025/actasciagron.v33i2.6120
• Mo M, Zhao L, Yang J, Song Y, Tu A, Liao K, Zhang J. Water and sediment runoff and soil moisture response to grass cover in sloping citrus land, Southern China. Soil Water Res. 2019;14:10-21. https://doi.org/10.17221/147/2017-SWR
  » https://doi.org/10.17221/147/2017-SWR
• Neves CSVJ, Tavares Filho J, Brito OR, Yamashita F, Tormem V, Fonseca ICB. Huerto de cítricos plantado con sistema de cero labranza y sistema convencional. Semin-Cienc Agrar. 2010:31:1263-74. https://doi.org/10.5433/1679-0359.2010v31n4Sup1p1263
  » https://doi.org/10.5433/1679-0359.2010v31n4Sup1p1263
• Novara A, Pulido M, Rodrigo-Comino J, Di Prima S, Smith P, Gristina L, Giménez-Morera A, Terol E, Salesa D, Keesstra S. Long-term organic farming on a citrus plantation results in soil organic carbon recovery. Cuad Investig Geogr. 2019;45:271-86. https://doi.org/10.18172/cig.3794
  » https://doi.org/10.18172/cig.3794
• Ojeda G, Patrício J, Navajas H, Comellas L, Alcañiz JM, Ortiz O, Marks E, Natal-Da-Luz T, Sousa JP. Effects of nonylphenols on soil microbial activity and water retention. Appl Soil Ecol. 2013;64:77-83. https://doi.org/10.1016/j.apsoil.2012.10.012
  » https://doi.org/10.1016/j.apsoil.2012.10.012
• Oliveira FÉR, Oliveira JM, Xavier FAS. Changes in soil organic carbon fractions in response to cover crops in an orange orchard. Rev Bras Cienc Solo. 2016;40:e0150105. https://doi.org/10.1590/18069657rbcs20150105
  » https://doi.org/10.1590/18069657rbcs20150105
• Reichert JM, Suzuki LEAS, Reinert DJ, Horn R, Håkansson I. Reference bulk density and critical degree-of-compactness for no-till crop production in subtropical highly weathered soils. Soil Till Res. 2009;102:242-54. https://doi.org/10.1016/j.still.2008.07.002
  » https://doi.org/10.1016/j.still.2008.07.002
• Santos RD, Lemos RC, Santos HG, Ker JC, Anjos LHC. Manual de descrição e coleta de solo no campo. 5. ed. Viçosa, MG: Sociedade Brasileira de Ciência do Solo; 2005.
• Silva AP, Kay BD, Perfect E. Characterization of the least limiting water range. Soil Sci Soc Am J. 1994;58:1775-81. https://doi.org/10.2136/sssaj1994.03615995005800060028x
  » https://doi.org/10.2136/sssaj1994.03615995005800060028x
• Silva JFG, Linhares AJS, Gonçalves WG, Costa KAP, Tormena CA, Silva BM, Oliveira GC, Severiano EC. Are the yield of sunflower and Paiaguas palisadegrass biomass influenced by soil physical quality? Soil Till Res. 2021;208:104873. https://doi.org/10.1016/j.still.2020.104873
  » https://doi.org/10.1016/j.still.2020.104873
• Silva RB, Lanças KP, Junior Masquetto B. Consolidômetro: Equipamento pneumáticoeletrônico para avaliação do estado de consolidação do solo. Rev Bras Cienc Solo. 2007;31:607-15. https://doi.org/10.1590/S0100-06832007000400001
  » https://doi.org/10.1590/S0100-06832007000400001
• Suzuki LEAS, Reichert JM, Reinert DJ, Lima CLR. Grau de compactação, propriedades físicas e rendimento de culturas em Latossolo e Argissolo. Pesq Agropec Bras. 2007;42:1159-67. https://doi.org/10.1590/S0100-204X2007000800013
  » https://doi.org/10.1590/S0100-204X2007000800013
• Thomaz EL, Fidalski J. Interrill erodibility of different sandy soils increases along a catena in the Caiuá Sandstone Formation. Rev Bras Cienc Solo. 2020;44:e0190064. https://doi.org/10.36783/18069657rbcs20190064
  » https://doi.org/10.36783/18069657rbcs20190064
• Tormena CA, Silva AP, Libardi PL. Soil physical quality of a Brazilian Oxisol under two tillage systems using the least limiting water range approach. Soil Till Res. 1999;52:223-32. https://doi.org/10.1016/S0167-1987(99)00086-0
  » https://doi.org/10.1016/S0167-1987(99)00086-0
• Turek ME, van Lier QJ, Armindo RA. Estimation and mapping of field capacity in Brazilian soils. Geoderma. 2020;376:114557. https://doi.org/10.1016/j.geoderma.2020.114557
  » https://doi.org/10.1016/j.geoderma.2020.114557
• Volsi B, Bordin I, Higashi GE, Telles TS. Economic profitability of crop rotation systems in the Caiuá sandstone area. Cienc Rural. 2020;50:e20190264. https://doi.org/10.1590/0103-8478cr20190264
  » https://doi.org/10.1590/0103-8478cr20190264

Edited by

Editors: José Miguel Reichert 0000-0001-9943-2898 and Gustavo Henrique Merten 0000-0003-0511-3218.

Publication Dates

• Publication in this collection
  27 June 2022
• Date of issue
  2022

History

• Received
  13 Jan 2022
• Accepted
  26 Apr 2022