Soil & Tillage Research 189 (2019) 140–147

Contents lists available at ScienceDirect

Soil & Tillage Research

journal homepage: www.elsevier.com/locate/still

Soil physical and mechanical attributes in response to successive harvests T

under sugarcane cultivation in Northeastern Brazil
⁎
Roberta Q. Cavalcantia, Mario M. Rolima, , Renato P. de Limaa, Uilka E. Tavaresa,
Elvira M.R. Pedrosaa, Igor F. Gomesb
a
Federal Rural University of Pernambuco, Rua Dom Manoel de Medeiros, Dois Irmãos, CEP: 52171-900, Recife, Pernambuco, Brazil
b
Federal University of Pernambuco, Rua Acadêmico Hélio Ramos, Cidade Universitária, CEP: 50670-420, Recife, Pernambuco, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: Soil degradation due to agricultural field traffic is a serious issue affecting soil functions in modern agriculture.
Soil compaction With sugarcane cultivation, the number of harvests through successive ratoons has become a challenge for the
Sugarcane expansion maintenance of soil quality in the topsoil and subsoil. The objective of this study was to evaluate physical and
Mechanized harvest mechanical responses of soil to successive harvests under sugarcane cultivation in an Ultisol of coastal tablelands
Ultisol
in Northeastern Brazil. Disturbed and undisturbed soil samples were collected from five sites at depths of
0–0.20 m (topsoil) and 0.20–0.40 m (subsoil), in which one, two, three or four harvests had taken place. In
addition, one site had been kept without cultivation for one year after a sugarcane crop. Organic carbon (OC),
maximum soil bulk density (BDmax), critical water content (wc), degree of compaction (DC), precompression
stress (σp), compression index (CI) and the angle of internal friction (φ) were measured and subjected to mul-
tivariate analyses for the different sugarcane ratoons. Soil compaction indicators and OC in the topsoil were
sensitive to the number of harvests, where compaction and OC increased with each ratoon sugarcane crop. The
increase in OC with successive harvests resulted in a decrease in the BDmax and an increase in the wc, which
suggest that optimum water content for compaction is moved to wetter zone due organic matter effects. The DC
increased considerably after the third harvest, and it was at a maximum level before the sugarcane renovation
(i.e. at the harvest immediately before the new seeding). In the subsoil, immediately below the plough layers,
where plough pan formation typically occurs in conventional tillage, considerable changes in the soil physical
and mechanical indicators are unlikely to occur due to the persistence of compacted layers. The increase in the
organic carbon content results in better soil physical condition in the soil surface, but the degree of compaction
may slightly increase due to frequency of traffic caused by the increase in the number of harvests, becoming
compaction more severe close to the sugarcane renovation.

1. Introduction presenting physical properties different to those observed below the

plough (Keller et al., 2014), in which the formation of a plough pan has
Brazil is one of the largest sugarcane producers in the world, with often been reported (Floyd, 1984; Keller et al., 2014).
extensive cultivated areas along the coastal tablelands in the Northeast The existence of these contrasting soil layers can yield different soil
region. Due to demand for bioenergy, there has been a recent increase responses (e.g. Keller et al., 2014) to sugarcane ratoons. Studies have
in sugarcane production, which has led to the expansion of cultivated reported that uncontrolled field traffic and tillage (e.g. Silva et al.,
areas (Cherubin et al., 2015). Studies have demonstrated that this ex- 2018) can cause soil physical degradation, influencing the soil structure
pansion has led to intensification of land use and caused changes in the and negatively affecting soil physical indicators (Castioni et al., 2018).
soil functions (e.g. Cherubin et al., 2015; Castioni et al., 2018) The result of these changes is the formation of compacted layers, which
Sugarcane is often cultivated in Brazil using conventional tillage have been reported to be a serious issue for soil functions in modern
systems. This includes the optional use of a subsoiler (0–0.40 m depth) agriculture (Nawaz et al., 2016).
in the compacted inter-rows and additional ploughing operations over Due to successive harvests in sugarcane areas, compaction can
the whole area at a depth of 0–0.20 m. Plough layers are loose, persist during the whole cycle before renovation (i.e. a new seeding).

⁎
Corresponding author.
E-mail addresses: [email protected], [email protected] (M.M. Rolim).

https://doi.org/10.1016/j.still.2019.01.006
Received 11 October 2018; Received in revised form 16 January 2019; Accepted 19 January 2019
0167-1987/ © 2019 Elsevier B.V. All rights reserved.
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

This involves a mean duration of five years between seeding and the dumpers and transported to the industrial processing unit. Conven-
renovation. During this five-year period, the cultivated areas are sub- tional loaders and dumpers were used throughout the experimental
jected to successive mechanized harvests (Lozano et al., 2013). Our fields without traffic control. Details of the vehicles used in the har-
knowledge regarding the evolution of soil physical and mechanical vesting operation can be found in Lozano et al. (2013).
indicators, such as degree of compaction, organic carbon content and
soil shear parameters due to these harvests, is limited and, therefore, 2.2. Sampling protocol and laboratorial tests
there is a lack of information on how the soil properties respond to
successive harvests in a ratoons crop. According to Lima et al. (2015), Four sampling locations were randomly established within each
these changes in the soil quality indicators can become more severe experimental site i.e., T0, T1, T2, T3 and T4 (numerals indicate the
because of the uncontrolled field traffic. number of harvests that had taken place see; Table 2), totalling four
Soil compaction can be measured by several soil compaction in- replicates per treatment. Samples were collected at the same time in all
dicators and soil compression properties (e.g. An et al., 2015; Lima areas. For T0, the experimental field was maintained without cultiva-
et al., 2018). The degree of compactness (DC), maximum bulk density tion for one year. At the time of sampling, the field was characterized
(BDmax) and critical (optimal) water content (wc) for soil compaction by bare soil. With the other treatments, the samples were collected after
are often obtained using the Proctor test and are useful parameters for sugarcane burning and harvesting. Disturbed and undisturbed soil
estimating the optimal water content for the occurrence of the max- samples were taken at each sampling location at depths of 0–0.20 m and
imum DC. The confined compression test is used to determine the 0.20–0.40 m.
precompression stress (σp) and compression index (CI), which can
provide information about the stress required for soil compaction, as 2.2.1. Disturbed soil samples
well as the rate of decrease in the soil volume due to external loads (e.g. The disturbed soil samples were used to determinate granulometric
by tyres). Additionally, knowledge of shear parameters is useful for a fractions (clay, silt and sand), organic carbon (OC), critical water
shear strength approach which is used to understand soil structure content (wc) and maximum bulk density (BDmax) by the Proctor test,
deterioration and tillage operations. For instance, Horn et al. (2003) and particle density (PD). The fractions of sand, silt and clay were de-
report that soil shearing which normally affects soil deformation and termined by the hydrometer method, as described by Grossman and
which result in no volume change may cause structure deterioration Reinsch (2002). The OC content was measured using the Walkley-Black
and homogenization of the pore system. method, and PD was determined using the volumetric-flask method
Compaction and shear strength parameters vary significantly with (EMBRAPA, 1997).
the soil bulk density (e.g. Zhang et al., 2001; Saffih-Hdadi et al., 2009; Proctor tests were performed in accordance with the Brazilian
Lima et al., 2018), water content (e.g. Saffih-Hdadi et al., 2009; Association of Technical Standards (ABNT) methodology (ABNT, 1990).
Schjønning and Lamandé, 2018) and organic carbon content (e.g. The soil was successively wetted and weighed in order to determine the
Braida et al., 2007; Pereira et al., 2007). Thus, changes in physical soil bulk density (BD) as a function of gravimetric water content (w).
parameters due to field traffic and soil management can result in The BDmax and wc were estimated at the maximum curvature point of
changes in the mechanical behavior of soil. the quadratic relationship between BD and w from the Proctor test.
Therefore, the objective of this study was to evaluate physical and
mechanical responses of soil to successive harvests under sugarcane 2.2.2. Undisturbed soil samples
cultivation in an Ultisol of coastal tablelands in Northeastern Brazil. In Undisturbed soil cores were sampled at depths of 0–0.20 m and
this context, we hypothesized that the successive harvests involved in 0.20–0.40 m using steel rings, 0.05 m in diameter and 0.05 m in height.
ratoon sugarcane cultivation over the years influence physical and Immediately after sampling, the soil cores were arranged and placed in
mechanical attributes of soil in the top- and subsoil due to the traffic. appropriate containers for moisture conservation and transportation. In
the laboratory, the cores were oven-dried at 105 °C for 24 h. BD was
2. Material and methods calculated from the weight of the oven-dried soil and the total volume
of the soil cores. Gravimetric water content (w) at sampling was then
2.1. Location and characterization of the experimental sites calculated as the difference between the wet weight of each soil core at
sampling and that after oven-drying. Degree of compactness (DC) was
The study was carried out at the Estação Experimental de Cana-de- calculated using the relationship between BD (from field) and the
açúcar do Carpina (EECAC-UFRPE) (Experimental Sugarcane Research BDmax.
Station), located in Carpina (7°51′S and 35°14′W), State of Pernambuco, Additionally, soil cores of 0.065 m diameter and 0.025 m height
Northeast Region, Brazil (Fig. 1), at an average altitude of 178 m. were sampled at depths of 0–0.20 m and 0.20–0.40 m for compression
Carpina has an annual mean rainfall of approximately 1400 mm and an testing. The soil cores were saturated by capillary action and thereafter
annual mean temperature of 24 °C. The soil studied is a sandy-loam each soil core was subjected to confined uniaxial compression tests in a
texture, classified as Yellow Argisol with a cohesive character according Bishop-type compression apparatus. Each of the following vertical
to EMBRAPA (2013). The term “cohesive” has been used in Brazil stresses was applied for two minutes in a sequential loading mode
(EMBRAPA, 2013) to distinguish soils with behavior similar to the (ABNT, 1990): 12.5, 25, 50, 100, 200, 400, 800 and 1600 kPa. Soil
Australian soils identified as hardsetting (see Giarola and Silva, 2002). displacement was measured after each loading step using a displace-
The Yellow Argisol corresponds to an Ultisol according to the soil ment transducer and the values used to calculate changes in soil vo-
classes of Soil Taxonomy (Soil Survey Staff, 2010). The basic physical lume. After the uniaxial compression tests, the cores were oven-dried at
characterisation of soil is presented in Table 1. 105 °C for 24 h. The BD was calculated from the weight of the oven-
For this study, five sites were chosen based on the number of har- dried soil and the total volume of the soil cores after each loading step.
vests that had taken place on the site (Table 2), with each harvest Precompression stress (σp) and the compression index (IC) were
corresponding to one year of cultivation. All experimental sites were calculated using the function sigmaP, available in the R package soil-
initially subjected to conventional tillage with a subsoiler at 0.40 m physics (Silva and Lima, 2015; R Core Team, 2017). The σp was de-
depth, followed by a disk harrow operation at 0.20 m depth. Soon after termined using the Pacheco Silva method, in accordance with ABNT
tilling, the soil was furrowed with a spacing of 1.0 m and a depth of (1990).
0.20 m; this is where the sugarcane was then seeded. At each harvest, Undisturbed soil blocks with dimensions of 0.20 × 0.20 × 0.20 m
the sugarcane was first burned, manually cut and then mechanically were sampled at depths of 0–0.20 m and 0.20–0.40 m using wooden
collected using conventional loaders. Stem sugarcane was placed on boxes. The blocks were immediately arranged and placed in

141
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

Fig. 1. A) Location of the Experimental Sugarcane Station of Carpina, Carpina, State of Pernambuco, Brazil. B) Location of the experimental sites with the treatments
T0, T1, T2, T3 and T4. T0: soil tilled and kept without cultivation for one year; T1: after one harvest; T2: after two harvests; T3: after three harvests; T4: after four
harvests. Source: Google Earth.

Table 1 appropriate containers for moisture conservation and transportation. In

Physical characterization of an Ultisol cultivated with sugarcane in different the laboratory, samples were extracted from the undisturbed soil blocks
ratoon crops. using square steel boxes of 0.060 m width and 0.025 m height. The
Treatments (T) Sand (g Silt (g Clay (g BD (Mg PD (Mg samples had been previously saturated by capillary action and were
kg−1) kg−1) kg−1) m−³) m−³) then subjected to direct shear tests. For the samples from 0–0.20 m
depth, normal stresses of 10, 30, 40 and 50 kPa were applied, whereas
Topsoil (0–0.20 m)
for the samples from 0.20–0.40 depth, the normal stresses of 55, 60, 65
T0 800 100 100 1.46 2.62
T1 730 120 160 1.44 2.62 and 70 kPa, were applied, simulating the levels of effective vertical
T2 680 130 190 1.43 2.63 geostatic stresses on the soil. The applied velocity in the test was
T3 750 110 140 1.43 2.59 0.30 mm min−1, thus characterizing the study as a fast direct shear test.
T4 770 100 130 1.60 2.58 At the end of the tests, the linear model of Mohr-Coulomb (Eq. 1) was
Subsoil (0.20–0.40 m)
T0 710 110 170 1.58 2.63
used to represent the shear stress (τ) as a function of the applied normal
T1 750 100 180 1.58 2.67 stress (σ).
T2 640 120 240 1.59 2.65
T3 690 80 230 1.55 2.65 τ = c + σ tan(φ) (1)
T4 720 100 190 1.56 2.64
Shear strength parameters, c (soil cohesion) and φ (angle of internal
PD: particle density; BD: bulk density; T0: soil tilled and kept without culti-
friction), were then extracted from the Mohr-Coulomb equation and
vation for one year; T1: after one harvest; T2: after two harvests; T3: after three
fitted to the experimental data for each treatment.
harvests; T4: after four harvests.

142
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

Table 2
Summary of the treatments (T) at sampling. Numerals indicate the number of harvests that had taken place.
Treatments (T) Characterization

T0 Treatment taken as reference. The soil was tilled and kept without cultivation for one year after sugarcane cultivation. At the time of soil sampling, the field was
characterized by bare soil.
T1 Experimental site under first ratoon crop. Soil tilled and planted, with occurrence of one harvest. Sugarcane productivity: 128.3 ton ha−1.
T2 Experimental site under second ratoon crop. Soil tilled and planted, with occurrence of two harvests. Sugarcane productivity: 86 ton ha−1.
T3 Experimental site under third ratoon crop. Soil tilled and planted, with occurrence of three harvests. Sugarcane productivity: 72.8 ton ha−1.
T4 Experimental site under fourth ratoon crop. Soil tilled and planted, with occurrence of four harvests. Sugarcane productivity: 60.6 ton ha−1.

2.3. Statistical analysis Thus, seven original variables (Fig. 4) were reduced to only two. Sta-
tistically, more than 90% of the experimental variability was main-
The data were subjected to descriptive statistical analysis, as well as tained for topsoil and subsoil. A theoretical threshold for scores
canonical discriminant analysis, in a multivariate approach with the (weights) > 0.4 (positive or negative), given by the red lines in Fig. 4,
intention of grouping and reducing original variables to canonical identifies the original variables with the greatest contribution within
variables. With this procedure, we aimed to reduce a large number of the canonical variables, i.e. the variables which were sensitive to the
variables in only one or two canonical variables that contain the soil successive harvests. It is possible to verify that in the topsoil the σp, DC,
physical and mechanical information of the original variables and that OC, wc (negative values), and the BDmax (positive value) presented
explain high experimental variability. Biplot graphs were used to ex- scores > 0.4 within CAN1 (Fig. 4). For CAN2, only the σp was shown to
amine the dispersion of the canonical scores associated with original be an important source of variation (i.e. score > 0.4) to discriminate
variables, as well as the impact of these variables on the discrimination the treatments. CAN1 for the subsoil shows that CI, CO, wc and BDmax
of the treatments. The mean values of the canonical variables for each had a score > 0.4, whereas OC had a score > 0.4 in CAN2 (Fig. 4B.2).
treatment were compared by 95% confidence spheres. All multivariate The dispersion of the scores for the original variables, which relates
analyses were performed through the candisc package, using R Software the changes in CAN1 and CAN2 to the treatments, is presented in Fig. 5.
(R Core Team, 2017). The spheres represent the average of CAN1 and CAN2 and the respective
95% confidence intervals for the treatments in the topsoil and subsoil.
3. Results T0 and T1 presented a positive average value for CAN1 and, therefore,
expressed the greatest values of BDmax (notice arrow size and direction).
Soil physical and mechanical indicators were modified over time in T4 presented the greatest negative average value of CAN1, expressing
the topsoil and subsoil due to the number of sugarcane ratoon harvests. the greatest values of σp, OC and wc, but mainly, higher DC values
The increase in the critical water content (wc) was continuous from T0 (Fig. 5A). In the subsoil, T0 presented a negative average value of
to T2, being approximately 0.04 g g−1 for topsoil and subsoil. The wc CAN1, with the greatest and lowest values of BDmax and OC, respec-
increased with number of harvests, reaching the highest values at T2 tively (Fig. 5B). T1, T2, T3 and T4 were statistically similar regarding
and remaining stable up to T4 (Fig. 2A), i.e. there were no considerable CAN1 (average close to 0; Fig. 5) and, therefore, presented physical and
changes from T2 to T4. mechanical attributes that were rather similar.
The DC was shown to have undergone changes only from T3 to T4 in
the topsoil. Changes in the DC were not verified from T0 to T2. On other 4. Discussion
hand, compaction continually increased after T3, reaching DC close to
85% at T4 (Fig. 2B). For the topsoil, compaction seems have been more Our results show that there was a continuous increase in OC in the
severe after three harvests (i.e. after T3). In the subsoil, there were no topsoil with the number of ratoon sugarcane harvests, which might
changes with the successive harvests. However, the DC was greater than have resulted the decrease in BDmax. BDmax was greater in T0 (bare soil)
80% at T0 and remained constant up to T4. Overall, the DC between and in the first years of the ratoon sugarcane crop, where the impact of
topsoil and subsoil were rather similar at T4. soil tillage should be more prominent and OC content was lower. This
The relative OC content of the topsoil and subsoil reversed from T2 indicates that OC reduced the BDmax, which was greater for the first
to T4, i.e. up to T2, OC was lower in the topsoil than the subsoil, after stages of sugarcane cultivation and decreased with successive harvests.
which, with T3 and T4, topsoil levels were higher than subsoil levels Marcolin and Klein (2011) found a linear and negative relationship
(Fig.2C). OC decreased in the subsoil and increased in the topsoil with between the BDmax and OC. They showed that the BDmax decreased with
successive harvests. OC content was maximum at T4 in the topsoil increasing OC, since the OC may promote a decrease in soil bulk density
(Fig. 2C). due to effects which result in decreased particle density (see Schjønning
The precompression stress (σp) was greater in the subsoil than in the et al., 2017), which occurs via a mass effect, expressed as a mixture
topsoil up to T2 (Fig. 3A), after which, at T3 and T4, values for topsoil ratio between organic and mineral soil components (Rühlmann et al.,
and subsoil were similar (c. 40 kPa). σp seems has been remained stable 2006). Changes in the BDmax associated with OC may be relevant from
in the subsoil with the number of harvests, but there was a considerable the point of view of soil compaction since, under the same compaction
increase in the topsoil from T0 to T4 (20 to 40 kPa). The CI was greater energy (i.e. Proctor test compaction energy), the soil could reach a
in the topsoil than in the subsoil (Fig. 3B). Overall, CI decreased with higher bulk density under conditions of lower OC. This would have a
the number of harvests, presenting the lower values at T4 (i.e., after different effect on the reduction of the pore space and soil aeration in a
four harvests). The angle of internal friction (φ) appeared not to have field subjected to traffic.
undergone changes, but it was greater in the subsoil (c. 40°) than in the While the increase in OC caused a decrease in BDmax, it resulted in
topsoil (c. 35°) (Fig. 3C). BDmax was rather variable, but decreased from an increase in the wc. This can occur due the high capacity of water
T0 to T4, i.e. with an increasing number of harvests (Fig. 3D). Overall, absorption by the organic matter, which may have promoted an in-
the BDmax was greater in the topsoil than in the subsoil (Fig. 3D). direct effect on water retention in the soil, as reported by Pereira et al.
Fig. 4 shows the impact of the original variables on the formation of (2007). A higher wc could lead the optimal water content for soil
two new variables, CAN1 and CAN2. It was possible to retain 92% of the compaction to be close to that of the soil wet zone (i.e. saturation),
experimental variability in CAN1 (77%) and CAN2 (15%) for the top- where the traffic is typically avoided due to known soil susceptibility to
soil, and 89% in CAN1 (77%) and CAN2 (12%) for the subsoil (Fig. 4). compaction (e.g. Lima et al., 2018). The results presented here show

143
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

Fig. 2. Changes in the critical water content (wc) (A), degree

of compactness (DC) (B) and organic carbon (OC) (C), as af-
fected by the number of harvests of ratoon crops (T0, T1, T2,
T3 and T4) in areas cultivated with sugarcane, at depths of
0–0.20 m (topsoil) and 0.20–0.40 m (subsoil). T0: soil tilled
and kept without cultivation for one year; T1: after one har-
vest; T2: after two harvests; T3: after three harvests; T4: after
four harvests. The bars represent the standard deviation for
each treatment.

Fig. 3. Changes in the precompression stress (σp) (A), com-

pression index (IC) (B), angle of internal friction (φ) (C), and
maximum soil bulk density (BDmax) (D), as affected by the
number of harvests of ratoon crops (T0, T1, T2, T3 and T4) in
areas cultivated with sugarcane at depths of 0–0.20 m (top-
soil) and 0.20-0.40 m (subsoil). T0: soil tilled and kept without
cultivation for one year; T1: after one harvest; T2: after two
harvests; T3: after three harvests; T4: after four harvests. The
bars represent the standard deviation for each treatment.

144
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

Fig. 4. Value of the scores of the original variables of the

CAN1 and CAN2 (canonical variables 1 and 2, respectively) at
depths of 0–0.20 m (topsoil; A) and 0.20–0.40 m (subsoil; B).
Red lines represent a theoretical threshold for those variables
with scores > 0.4 (positive or negative). σp: precompression
stress (kPa); φ: angle of internal friction (°); IC: compression
index (-); BDmax: maximum soil bulk density (Mg m−3); wc:
critical water content (g g-1); OC: organic carbon (g kg-1); DC:
degree of compactness (%).

that OC management could be important for reducing the impact of sugarcane harvest and transportation can vary from 150 to 600 kPa.
agricultural field traffic on the mass-volume ratio, although the effects They argue that the logistics of the sugarcane harvest allows the pas-
in the subsoil are less relevant than in the topsoil because of lower OC sage of the machinery across the cultivated field. Topsoil is rather
content (Fig. 2C). susceptible to the stress applied by tyres because this layer is close to
The DC in the topsoil increased with number of harvests, whereas the tyre-soil interface, where the highest stresses are recorded (Keller,
the level was kept practically constant in the subsoil. Tillage effects 2005; Schjønning et al., 2008). According to Castioni et al. (2018), the
might have prevailed in the first years of sugarcane cultivation, making heavy and uncontrolled traffic in sugarcane fields can cause severe
the soil loose and susceptible to vertical displacement. Additionally, the damage to the soil structure, corroborating the increase in the DC ob-
results show that compaction in the topsoil (here assumed to be the DC) served for topsoil along with the number of harvests.
increased along with the number of harvests, and this may have oc- In conventional tillage systems, the most severe compaction occurs
curred due to the impact of field traffic. Lozano et al. (2013) reported immediately below the plough layer in the form of a plough pan
that the stress applied to the topsoil by several machines used in (Olesen and Munkholm, 2007). Plough pans can persist in the subsoil

Fig. 5. Dispersion of the canonical scores associated with

original variables and 95% confidence spheres for the average
of the CAN1 and CAN2 (canonical variables 1 and 2, respec-
tively) of the treatments T0, T1, T2, T3 and T4 at depths of
0–0.20 m and 0.20–0.40 m. σp: precompression stress (kPa); φ:
angle of internal friction (°); IC: compression index (-); BDmax:
maximum soil bulk density (Mg m−3); wc: critical water
content (g g-1); OC: organic carbon (g kg-1); DC: degree of
compactness (%). T0: soil tilled and kept without cultivation
for one year; T1: after one harvest; T2: after two harvests; T3:
after three harvests; T4: after four harvests.

145
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

for a long-time in case of no adoption of mechanical practice to alle- mechanical indicators are unlikely to occur due to the persistence of
viate soil compaction. Compaction in the subsoil in this study was de- compacted layers.
monstrated by the higher DC values, mainly when they were compared The increase in the organic carbon content results in better soil
to topsoil measurements (Fig. 2C). The DC was not modified by the physical condition in the soil surface, but the degree of compaction may
number of harvests (i.e. from T0 to T4), showing that a residual plough slightly increase due to frequency of traffic caused by the increase in the
pan caused by several years of cultivation, could be present in the number of harvests, becoming compaction more severe close to the
subsoil. Nunes et al. (2015) verified experimentally that the effect of sugarcane renovation.
mechanical practices did not promote persistent improvements in the
soil structure and that soil compaction could return even in short time Acknowledgments
due to the machinery traffic. The use of a subsoiler before sugarcane
seeding was not enough to alleviate compaction in the subsoil. It ap- We would like to thank the Coordenação de Aperfeiçoamento de
pears that the residual plough pan persisted after sugarcane renovation, Pessoal de Nível Superior (CAPES, Brazil) and the Conselho Nacional de
resulting in similar DC levels with all treatments. According to Floyd Desenvolvimento Cientifico e Tecnológico (CNPq, Brazil) for granting
(1984) and Keller et al. (2014), plough pans developed at depths be- scholarships and financial support.
tween 20 and 30 cm, which was the layer adopted to characterize the
subsoil within this study. References
The values of σp slightly increased in the topsoil with the number of
harvests. However, similar to the DC, the σp showed little variation in ABNT - Associação Brasileira de Normas Técnicas, 1990. NBR 12007: Ensaio de aden-
the subsoil. The σp has been reported to vary with the bulk density (e.g. samento unidimensional. ABNT, Rio de Janeiro 13p.
An, J., Zhang, Y., Yu, N., 2015. Quantifying the effect of soil physical properties on the
Saffih-Hdadi et al., 2009; Lima et al., 2015), that is with compaction. compressive characteristics of two arable soils using uniaxial compression tests. Soil
Saffih-Hdadi et al. (2009) described a linear model where there was an Tillage Res. 145, 216–223.
increase in the σp with the bulk density. Therefore, the increase in the Braida, J.A., Reichert, J.M., Reinert, D.J., Soares, J.M.D., 2007. Coesão e atrito interno
associados aos teores de carbono orgânico e de água de um solo franco arenoso.
σp observed in the topsoil could be attributed to an increase in the DC, Ciência Rural 37, 1646–1653.
corroborating findings of Saffih-Hdadi et al. (2009) and Lima et al. Castioni, G.A., Cherubin, M.R., Menandro, L.M.S., Sanches, G.M., Bordonal, R.O.,
(2015). The results indicate that σp becomes greater with the number of Barbosa, L.C., Franco, H.C.J., Carvalho, J.L.N., 2018. Soil physical quality response to
sugarcane straw removal in Brazil: a multiapproach assessment. Soil Tillage Res. 184,
harvests as a result of the traffic.
301–309.
The CI and φ did not change with the number of harvests, but there Cherubin, M.R., Karlen, D.L., Francon, A.L.C., Tormena, C.A., Cerri, C.E.P., Davies, C.A.,
were differences between the topsoil (0.17) and subsoil (0.23). CI in- Cerri, C.C., 2015. Soil physical quality response to sugarcane expansion in Brazil.
Geoderma 267, 156–168.
dicates the rate of change in the soil volume with the applied stress, i.e.
EMBRAPA - Empresa Brasileira de Pesquisa Agropecuária, 2013. Sistema brasileiro de
the increase, in the bulk density (or decrease in the soil volume) with classificação de solos, 3 ed. Brasília. 353p.
the applied stress. Our measurements show that the CI was higher in the EMBRAPA- Empresa Brasileira de Pesquisa Agropecuária, 1997. Manual de métodos de
topsoil. Imhoff et al. (2004); Saffih-Hdadi et al. (2009) and Lima et al. análises de solo. 2.ed. Ministério da Agricultura e do Abastecimento, Rio de Janeiro
212 p.
(2018) demonstrated that the CI decreased with the increase in soil bulk Floyd, C.N., 1984. Model experiment on the effect of a plough pan on crop yield under
density. Therefore, it is likely that the higher CI in the topsoil was as- differing conditions of soil moisture availability. Soil Tillage Res. 4, 175–189.
sociated with the lower DC. Giarola, N.F.B., Silva, A.P., 2002. Conceitos sobre solos coesos e hardsetting. Sci. Agric.
59, 613–620.
Shear proprieties of agricultural soil typically vary with water Grossman, R.B., Reinsch, T.G., 2002. Bulk density and linear extensibility. In: Dane, J.H.,
content and bulk density (e.g., Zhang et al., 2001; Braida et al., 2007). Topp, G.C. (Eds.), Methods of Soil Analysis. Physical Methods, IV. Soil Science
In this study, water content was constant, while the bulk density was Society of America, Madison, pp. 201–225.
Horn, R., Way, T., Rostek, J., 2003. Effect of repeated tractor wheeling on stress/strain
dependent on the treatments. DC was greater in the subsoil than in the properties and consequences on physical properties in structured arable soils. Soil
topsoil and corroborates the greater φ measured. The study of Braida Tillage Res. 73, 101–106.
et al. (2007) involving agricultural soils reported that the φ positively Imhoff, S., Silva, A.P., Fallow, D., 2004. Susceptibility to compaction, load support ca-
pacity, and soil compressibility of Hapludox. Soil Sci. Soc. Am. J. 68, 17–24.
correlated with the shear stress, i.e. the stress required for the soil to Keller, T., 2005. A model to predict the contact area and the distribution of vertical stress
rupture. Shear stress was reported to be greater with the increase in the below agricultural tyres from readily available tyre parameters. Biosyst. Eng. 92,
soil bulk density (or degree of compactness) (Silva and Carvalho, 2007). 85–96.
Keller, T., Berli, M., Ruiz, S., Lamandé, M., Arvidsson, J., Schjønning, P., Selvadurai,
The number of harvests did not cause changes in the φ, but due to the
A.P.S., 2014. Transmission of vertical soil stress under agricultural tyres: comparing
compaction aspect (e.g. plough pan), it is expected that the subsoil measurements with simulations. Soil Tillage Res. 140, 106–117.
would present higher shear strength than topsoil. On a practical effect, Lima, R.P., Rolim, M.M., Oliveira, V.S., Silva, A.R., Pedrosa, E.M.R., Ferreira, R.L.C.,
the lower shear strength observed in the topsoil could favor tillage 2015. Load-bearing capacity and its relationships with the physical and mechanical
attributes of cohesive soil. J Terramechanics 58, 51–58.
operations. However, lower shear strength could be a problem for soil Lima, R.P., Silva, A.P., Giarola, N.F., Silva, A.R., Rolim, M.M., Keller, T., 2018. Impact of
structure, which becomes susceptible to rupture even under low applied initial bulk density and matric suction on compressive properties of two Oxisols
stress. under no-till. Soil Till Res. 175, 168–177.
Lozano, N., Rolim, M.M., Oliveira, V.S., Tavares, U.E., Pedrosa, E.M.R., 2013. Evaluation
of soil compaction by modeling field vehicle traffic with SoilFlex during sugarcane
5. Conclusions harvest. Soil Till Res. 129, 61–68.
Marcolin, C.D., Klein, V.A., 2011. Determinação da densidade relativa do solo por uma
função de pedotransferência para a densidade do solo máxima. Acta Sci. Agron. 33,
Soil compaction indicators and organic carbon in the topsoil are 349–354.
sensitive to the number of harvests in sugarcane cultivation, where Nawaz, M.F., Bourrié, G., Trolard, F., Ranger, J., Gul, S., Niazi, N.K., 2016. Early detec-
compaction and organic carbon increase with ratoon sugarcane. The tion of the effects of compaction in forested soils: evidence from selective extraction
techniques. J. Soils Sediments 16, 2223–2233.
increase in organic carbon over successive harvests results in a decrease Nunes, M.R., Dernadin, J.E., Pauletto, E.A., Faganello, A., Pinto, L.F.S., 2015. Effect of soil
in the maximum soil bulk density and an increase in the optimum water chiseling on soil structure and root growth for a clayey soil under no-tillage.
content, which suggests that optimum water content for compaction is Geoderma 259–260, 149–155.
Olesen, J.E., Munkholm, L.J., 2007. Subsoil loosening in a crop rotation for organic
moved to wetter zone due organic matter effects. The degree of com-
farming eliminated plough pan with mixed effects on crop yield. Soil Tillage Res. 94,
paction increased considerably after the third harvest, and reached a 376–385.
maximum level before sugarcane renovation (i.e. at the harvest im- Pereira, J.O., Defossez, P., Richard, G., 2007. Soil susceptibility to compaction by
mediately before the new seeding). In the subsoil, immediately below wheeling as a function of some properties of a silty soil as affected by the tillage
system. Eur J Soil Sci. 58, 34–44.
the plough layers, where plough pans are typically reported with con- R Core Team, 2017. R: A Language and Environment for Statistical Computing [Internet].
ventional tillage, considerable changes in the soil physical and

146
R.Q. Cavalcanti et al. Soil & Tillage Research 189 (2019) 140–147

R Foundation for statistical computing, Vienna, Austria. Available at: http://www.R- density from clay and soil organic matter contents. Geoderma 286, 83–87.
project.org. Accessed 19.09.17. Silva, A.J.N., Carvalho, F.G., 2007. Coesão e resistência ao cisalhamento relacionadas a
Rühlmann, J., Korschens, M., Graefe, J., 2006. A new approach to calculate the particle atributos físicos e químicos de um Latossolo Amarelo de Tabuleiro Costeiro. R. Bras.
density of soils considering properties of the soil organic matter and the mineral Ci. Solo 31, 853–862.
matrix. Geoderma 130, 272–283. Silva, A.R., Lima, R.P., 2015. soilphysics: an R package to determine soil preconsolidation
Saffih-Hdadi, K., Defossez, P., Richard, G., Cui, Y.J., Tang, A.M., Chaplain, V.A., 2009. pressure. Comput Geosci. 84, 54–60.
Method for predicting soil susceptibility to the compaction of surface layers as a Silva, R.P., Rolim, M.M., Gomes, I.F., Pedrosa, E.M.R., Tavares, U.E., Santos, A.N., 2018.
function of water content and bulk density. Soil Tillage Res. 105, 96–103. Numerical modeling of soil compaction in a sugarcane crop using the finite element
Schjønning, P., Lamandé, M., 2018. Models for prediction of soil precompression stress method. Soil Tillage Res. 181, 1–10.
from readily available soil properties. Geoderma 320, 115–125. Soil Survey Staff, 2010. Keys to Soil Taxonomy, 11th ed. United States Department of
Schjønning, P., Lamande, M., Tøgersen, F.A., Arvidsson, J., Keller, T., 2008. Modelling Agriculture, Natural Resources Conservation Service, Washington.
effects of tyre inflation pressure on the stress distribution near the soil-tyre interface. Zhang, B., Zhao, Q.G., Horn, R., Baumgartl, T., 2001. Shear strength of surface soil as
Biosyst. Eng. 99, 119–133. affected by soil bulk density and soil water content. Soil Till. Res. 59, 97–106.
Schjønning, P., McBride, R.A., Keller, T., Obour, P.B., 2017. Predicting soil particle

147