Critical limits of soil physical attributes for corn and black oat in a Xanthic Hapludox

- Soil compaction can decrease potential crop yield and is frequently found in no-till cultivation systems, but there is little information on the critical limits of soil physical properties in these conditions, especially in subtropical soils. The objectives of this study were to evaluate the in ﬂ uence of tractor traf ﬁ c on soil physical properties and identi ﬁ es critical limits of soil bulk density, soil compaction degree, macro, and microporosity on crop yield of corn and black oat assessed in situ in a Xanthic Hapludox in Southern Brazil. Tractor machine traf ﬁ c (0, 2, 5 and 20 passes) was established in no-till soil to create different soil compaction levels. Undisturbed soil samples were collected at depths of 0.00-0.05, 0.07-0.12 and 0.17-0.22 m just after tractor traf ﬁ c in two moments: before corn cultivation, and then before black oat cultivation. Under the conditions of this experiment, the 10% higher yields of corn and black oats were obtained with macroporosity above 0.10 m 3 m -3 and 0.13 m 3 m -3 , respectively. Similarly, the soil compactness degrees of 90% and 86% were considered critical for the corn and black oat, respectively. In addition, soil compaction by machine traf ﬁ c reduced maize yield up to 33% and black oat biomass yield up to 44%. Our results demonstrate that growers should be aware of the consequences of machine traf ﬁ c on soil properties, as it can severely compromise crop yields, especially of corn and black oats cultivated in very clayey soil like the one evaluated in this study.


INTRODUCTION
Soil compaction especially occurs in notill cultivation because there is intense traffic of agricultural machines (FERREIRA et al., 2020;TRENTIN et al., 2018) or inadequate cropping systems (SOMASUNDARAM et al., 2018) , which affects soil structure and has a direct effect on crop yield (LIMA et al., 2018;MARINS et al., 2018;REICHERT et al., 2009).
The compacted soil impairs root growth, reduces hydraulic conductivity, especially the saturated one, and reduces air-fi lled porosity (REICHARDT; TIMM, 2020). Morphological and physiological plant attributes are affected by the aforementioned soil physical attributes, such as stunting growth, reducing plant height, and implementing a shallow root system (SHAH et al., 2017); it also decreases water and nutrient absorption, thus decreasing crop yields (COLOMBI; KELLER, 2019;MORAES et al., 2020;RIVERA et al., 2019).
Growers occasionally carry out mechanical chiseling of fi elds under a no-till system to mitigate soil compaction, which is not an indicated practice (MORAES et al., 2020), and some Brazilian farmers have been chiseling the soil under a no-till system every 2 to 5 years without considering soil physical attributes. However, it is prudent to be sure that the soil is compacted to chisel the soil, and the soil physical properties and critical limits should be assessed and established.
The machine traffic affects the soil degree of compactness, which in turn affects the total porosity, size, and dynamics of soil pores. Larger pores are represented by macroporosity and are the most important for soil aeration and the speed of water flow, while smaller pores, represented by microporosity, represent water which is strongly retained in the soil (REICHARDT; TIMM, 2020). Soil physical properties which are influenced by soil management and can be used as indicators of soil compaction include soil bulk density (BD) (MORAES et al., 2016;REICHERT et al., 2009;TRENTIN et al., 2018) and soil porosity (macroporosity and microporosity) (BARBOSA et al., 2019;MORAES et al., 2016). Despite the importance of these topics, the relationship between porosity, especially microporosity, and plant growth is little reported (MORAES et al., 2020). Although BD is one of the attributes more used to evaluate soil compaction (GUBIANI; REINERT; REICHERT, 2014), their values are dependent on particle size and organic matter content (MARCOLIN; KLEIN, 2011).
Accordingly, the use of soil degree of compactness (DC) represents the relationship between BD and the reference bulk density (BD ref ) obtained by Proctor test (MARCOLIN; KLEIN, 2011;OLIVEIRA et al., 2016), and is considered a useful indicator of soil physical condition as it is a way of normalizing soil density in relation to texture (KOUREH et al., 2020;OLIVEIRA et al., 2016;ROSSETTI;CENTURION, 2015;SUZUKI et al., 2007). Root growth is restricted when BD exceeds BD ref HÅKANSSON, 2010). The degree of compactness, like matric water tension, is related to the air-fi lled porosity and limits of penetration resistance, in addition to being similar in different soils, which makes it a useful parameter in studying the biological effects of soil compaction LIPIEC, 2000).
Critical values of DC and macroporosity depend on the soil texture . The macroporosity reference values found in literature vary from 0.08 to 0.14 m 3 m -3 (CARTER, 1990) and minimum values in subtropical highly weathered soils are reported as 0.10 m 3 m -3 to avoid compromising root growth (REICHERT; SUZUKI; REINERT, 2007); even though soil aeration is a dynamic process, root depth, soil type and soil respiration rate determine the air-fi lled porosity requirements (BARTHOLOMEUS et al., 2008). The lower limit of macroporosity varies according to crop root growth capacity; as the compaction impairs root growth, the soil area available for plant exploitation and root respiration impairs water relationships and nutrient availability (REICHARDT; TIMM, 2020). The macropore continuity in very clayey soils is greater and the reference value less than 0.10 m 3 m -3 would be enough to ensure aeration for the root system (REICHERT et al., 2009). The optimum DC for soybean cultivated in Oxisol was 86% , while the maximum yield for barley was found with DC between 77 and 84% (CARTER, 1990), indicating that it is considered a non-compacted range.
In view of the above, the objectives of this study were to evaluate the influence of tractor traffic on the soil physical properties of an oxisol and identify critical limits of soil bulk density, soil degree of compactness, macro and microporosity on crop yield of corn and black oat assessed in situ in a no-till system in Southern Brazil.

Site description, experimental design, and treatments
The experiment was carried out in Guarapuava, Parana, Brazil (25º 23'S and 51º 27' W, 1041 m above sea level), under humid subtropical mesothermic climate (Cfb) and the topography is smooth to gently undulated. The soil of the experimental site is a Xanthic Hapludox with a very clayey texture. The soil granulometric analysis indicated 748, 215, and 37 g kg -1 of clay, silt, and sand, respectively. Critical limits of soil physical attributes for corn and black oat in a Xanthic Hapludox The experimental design was of randomized blocks arrangement, with 4 treatments and 6 replications, and the plot size was 80 m 2 (5 m × 16 m). The treatments consisted of the following soil compaction by tractor traffi c throughout the plot: control (without machine traffi c), 2 passes, 5 passes and 20 passes. Soil compaction in the experimental area was carried out on October 27, 2016, two days after a precipitation period of around 70 mm. The average soil moisture was 0.43 m³ m -³ at that moment, close to the humidity that the micropores were fi lled with water and the macropores were drained ( Table 2). The tractor used to generate different levels of soil compaction was a John Deere 7515 model with 140 hp power, weighted with ¾ of the tire volume with water, plus six front weights of 50 kg each and four rear weights of 75 kg each, with a total weight of 6 Mg.

Sampling and evaluation of soil physical properties
First, undisturbed soil samples were collected at depths of 0.00-0.05, 0.07-0.12 and 0.17-0.22 m in two moments: before corn cultivation on November 7, 2016 (10 days after soil compaction), and before black oat cultivation on June 22, 2017 (238 days after soil compaction), in order to determine the critical limits of soil physical properties. Thus, the soil bulk density (BD), as well as the macro and microporosity (EMBRAPA, 2011) were determined using undisturbed soil samples. The undisturbed soil samples were subsequently subjected to the tension table at -6 kPa after being saturated for 48 hours to determine macro and microporosity. Soil total porosity was the sum of macro and microporosity. Afterward, the soil samples were oven-dried at 105 ºC until constant weight to determine the BD.
Next, a normal Proctor test was used (MARCOLIN; KLEIN, 2011) to determine the reference bulk density (BD ref ). The test consisted of using an impact resulting from a fi xed load in soil samples submitted to different humidity, resulting in a density curve achieved as a function of humidity. A 2.5 kg socket, a cylindrical metallic mold (10 cm in diameter and 12.73 cm in height) and a complementary metal ring were used to simulate the impact load. The BD ref was determined in the same layers as the undisturbed samples (0.00-0.05, 0.07-0.12, and 0.17-0.22 m). The BD values of each layer and treatment were divided by the BD ref of each studied layer to determine soil degree of compactness (DC).

Management and evaluations of corn and black oat
After the soil compaction process, corn was sown on November 8, 2016, using a conventional hybrid DKB 290 with a seeding rate of 7.5 seeds m -2 and a row spacing of 0.90 m. Basic fertilization consisted of 26 kg ha -1 of N (urea), 90 kg ha -1 of P 2 O 5 (triple superphosphate), and 51 kg ha -1 of K 2 O (potassium chloride) and a top-dress fertilization with 100 kg ha -1 of N (urea) was c arried out 20 days after sowing. Preventive management of pests, diseases and weeds was properly carried out. Then, an area of 10 m 2 per plot was harvested at full maturity to determine corn grain yield and grain moisture was adjusted to 130 g kg -1 .
Black oat (Avena strigosa Schreb.) cv IAPAR 61 was sown on July 22, 2017 in the winter, with a seeding rate of 360 seeds m -2 and a row spacing of 0.17 m. Top-dress fertilization was carried out after 20 days with 100 kg ha -1 of N (urea). Then, an area of 0.5 m -2 per plot was cut at its full bloom stage at 0.01 m from soil surface and oven-dried at 60 ºC to evaluate black oat dry matter.

Statistical analysis
The data were subjected to the normality and homogeneity test of the residuals. The data were subjected to analysis of variance (ANOVA) under a factorial scheme (4 × 3) to relate machine traffi c and soil depths, and each sampling moment was independently evaluated. Corn grain yield and black oat dry matter yield were submitted to one-way ANOVA. The data were classifi ed by the Tukey test at the 5% signifi cance level when ANOVA showed a signifi cant difference (α = 0.05).
A regression study was carried out among the soil properties, relative corn yield and relative dry matter of black oat to determine the critical limits of soil physical properties. The relative crop yield was defi ned as the relationship between the crop yield and the average of the best crop yield among soil compaction treatments. The ideal limit of the physical properties was delimited at 90% of the relative corn crop yield and black oat dry matter using equations of fi rst or second degree, and using the coeffi cient of determination as criteria. The critical limit criteria adopted in this study is estimated within the transition zone, associated with a reduction in growth or yield in 10%, which is also used in soil fertility (NÚCLEO ESTADUAL PARANÁ -SOCIEDADE BRASILEIRA DE CIÊNCIA DO SOLO, 2017).

RESULTS AND DISCUSSION
The reference bulk density (BD ref ) obtained by Proctor test was 1.19, 1.20 and 1.21 Mg m -3 for the layers of 0.00-0.05, 0.07-0.12 and 0.17-0.22 m, respectively. The soil moisture content in BD max ranged from 0.32 to 0.36 kg kg -1 . There was a significant difference between the soil compaction treatments by machine traffic for the BD, DC, macro, and microporosity soil properties (Table 1). Total porosity was not different between the treatments.
The soil compaction was verifi ed at 0.00 until 0.22 m depth at the highest soil compaction level (20 passes of tractor traffic), as evidenced in the BD and DC values.
The corn yield was lower with 20 passes of the machine traffi c, while there was no difference among the *Uppercase letter on the line and lowercase letter in the column classify the means by Tukey's test (p < 0.05). ns: not signifi cant. The statistical comparison is within each crop * Lowercase letter in the column classify the averages by Tukey's test (p < 0.05) other treatments ( Table 2). The corn yields in the treatment with 20 passes were 34%, 28% and 34% lower than 0, 2, and 5 passes of the machine traffi c, respectively. The black oat dry matter was higher without soil compaction (0 passes) compared to the other treatments, which did not differ from each other. The black oat dry matter in the treatment with 0 passes were 33%, 50% and 78% higher than 2, 5, and 20 passes of the machine traffi c, respectively ( The relationship between DC with corn yield presented distinct behavior in the three layers, in which the 0.00-0.05 m layer obtained the critical DC limit of 90% ( Figure 1A), equivalent to BD limit of 1.07 Mg m -3 .
There was a quadratic fi t in the 0.07-0.12 m layers between the relative corn yield with DC, which was between 77 to 90% ( Figure 1C), equivalent to BD between 0.93 to 1.08 Mg m -3 . The ideal DC range in the 0.17-0.22 m layer was between 81 to 90% ( Figure 1E), equivalent to BD between 0.98 to 1.08 Mg m -3 . The most favorable crop performance of around 85% of degree of compactness was also reported by Håkansson and Lipiec (2000). There is quadratic fi t in deep soil layers because if the soil degree of compactness is high (critical superior boundary), there is higher water infi ltration, which increases the susceptibility of water defi cit by plants. If Critical limits of soil physical attributes for corn and black oat in a Xanthic Hapludox the soil degree of compactness is low (critical inferior boundary), it is evidence of low macroporosity and higher microporosity at a level which impairs root growth, with similar observations being reported by Carter (1990).
The external forces responsible for soil compaction (i.e. machine traffi c) mainly affect macroporosity, which in turn affects water infi ltration into the soil by reducing hydraulic conductivity, as well as reducing aeration (REICHARDT; TIMM, 2020).
Macroporosity and total porosity in the 0.00-0.05 m layer had a positive linear fi t compared with relative corn yield, while microporosity had a negative linear fi t with relative corn yield. Based on the 90% of relative corn yield, the ideal macroporosity in the 0.00-0.05 m layer ranged between 0.104 and 0.261 m 3 m -3 , the ideal microporosity was below 0.482 m 3 m -3 and the ideal total porosity was above 0.604 m 3 m -3 (Figure 2A).
Microporosity in the 0.07 to 0.12 m layer had a quadratic fi t compared with relative corn yield, while microporosity had a negative linear fi t and total porosity had a positive linear fi t. Based on the 90% of relative corn yield, the ideal macroporosity in the 0.07 to 0.12 m layer ranged between 0.122 and 0.215 m 3 m -3 , the ideal microporosity was below 0.458 m 3 m -3 and the ideal total porosity was above 0.616 m 3 m -3 ( Figure 2B).
Macroporosity in the 0.17-0.22 m layer had a positive linear fi t compared with relative corn yield, microporosity had a negative linear fi t and total porosity had no signifi cant linear nor quadratic fi t. Based on the 90% of relative corn yield, the ideal macroporosity in the 0.17-0.22 m layer ranged between 0.090 and 0.189 m 3 m -3 and the ideal microporosity was below 0.512 m 3 m -3 ( Figure 2C).
The relationship between BD or DC with relative black oat dry matter yield presented not signifi cant relationship in the 0.00-0.05 m ( Figure 1B) and 0.17-0.22 m layers ( Figure 1F), while there was a negative linear fi t in the 0.07-0.17 m layer ( Figure 1D). The 0.07-0.17 m layer presented a critical BD limit of 1.02 Mg m -3 and a critical DC of 86% ( Figure 1D).
The relationship between porosity with relative black oat dry matter yield presented no signifi cant relationship in the 0.00-0.05 m ( Figure 3A and B) and 0.17-0.22 m layers ( Figure 3E and F), while there was a negative linear fi t in the 0.07-0.17 m layer ( Figure 3C and D).
Based on the 90% of relative black oat dry matter, the ideal macroporosity in the 0.07-0.17 m layer is above 0.13 m 3 m -3 and the ideal total porosity is above 0.61 m 3 m -3 .
The soil compaction carried out by machine traffi c modifi ed soil physical properties, increased BD, DC and microporosity, and decreased the macroporosity to a depth of 0.22 m. The increase in BD and decrease macroporosity in compacted soils is commonly reported in different soil types (CARTER, 1990;LIMA et al., 2018;POTT et al., 2019;REICHERT et al., 2009). The critical BD to consider the Xanthic Hapludox evaluated in our study as a compacted soil ranged from 1.07 to 1.09 Mg m -3 for corn and 1.03 Mg m -3 for black oat. Another study on this same soil type considered non-compacted soil with BD greater than 1.30 Mg m -3 (ANDREOLLA et al., 2015) taking into account the references of Reichert et al. (2009). The lower BD values found in our study reinforce the need to assess BD ref , in which they ranged from 1.19 to 1.21 Mg m -3 . These low BD values indicate the importance of using DC to determine whether the soil is compacted or not (KOUREH et al., 2020;MARCOLIN;KLEIN, 2011;OLIVEIRA et al., 2016;ROSSETTI;CENTURION, 2015).
The DC of 90% resulted in a potential relative yield below 90% for corn ( Figure 1A, C, and E). Oliveira et al.
(2016) verifi ed a reduction by 50% of root length when the soil reached a DC of 98%. Suzuki et al. (2007) found that the optimum DC for soybean cultivation in Oxisol was 86%. Carter (1990) obtained a maximum yield of barley in DC with values between 77 and 84%, indicating that it was considered a non-compacted range. Sá et al. (2016) considered a DC value of 88% as critical for the growth of sugarcane roots. Silva, Albuquerque and Costa (2014) studying soybean growth in mini-lysimeter, found that a critical value of BD occurs when it was above 1.14 Mg m -3 , which corresponds to 95% of DC.
Black oat roots penetrate the compacted soil with no apparent deviation (BURR-HERSEY et al., 2017). In our study, black oat presented 90% of dry matter yield when the DC was greater than 86%. Pott et al. (2019) also verifi ed the response of dry matter of black oat due to different soil physical properties.
The increased tolerance of crops to compacted soils can be mitigated through genetic improvement and with management practices that recovery soil properties, such as porosity (COLOMBI; KELLER, 2019).
The soil macroporosity in our study was a good indicator of soil physical quality of black oat yield. Although signifi cant, this is a static soil property, which may not represent the temporal variability of factors which directly affect crop growth and development. According to Colombi and Keller (2019), root growth is decreased in compacted soil due to high soil penetration resistance and due to low oxygen concentration in soil air caused by reduced fl uid transport capability.
In this study, increasing the intensity of the machine traffic reduced the macroporosity, which consequently increased the microporosity. It was observed that the increase in microporosity occurred in a small range, and that there was a negative relationship with yield in the largest microporosities ( Figure 3). This negative relationship evidences that the greater microporosity may be related to the physical impediment of root growth provided by greater soil compaction. The compaction treatments from machine traffi c were installed before sowing the corn. The soil physical attributes were evaluated and then the corn was sown; the physical attributes of the soil were evaluated again, and the black oats were sown. The soil disturbance caused by sowing corn possibly caused the non-effect of the machine traffi c treatments on the macroporosity in the 0.00-0.05 m surface layer (Figure 3a). The non-effect of macroporosity in the lower subsurface layer evaluated (0.17-0.22 m) also evidences that the root growth of corn mitigated the compaction effects in the layers (Figure 3c), which can be an effect of "biological decompression" by corn roots. This information was also evidenced by bulk density (Table 1), where the amplitude of bulk density between machine traffi c treatments within the 0.17-0.22 m layer were clearly reduced when evaluated before black oat when compared to the evaluation before corn.
In order to reach at least 90% of maximum relative crop yield in the Xanthic Hapludox evaluated in the 0.00 to 0.22 m layers, a minimum macroporosity from 0.10 to 0.12 m 3 m -3 was necessary for corn cultivation, and at least 0.13 m 3 m -3 for black oat cultivation (Figures 2 and 3). It was reported that the macroporosity of 0.10 m 3 m -3 in some very clayey soils from subtropical environments will not necessarily be limiting for root growth in no-tillage (REICHERT; SUZUKI; REINERT, 2007;REICHERT et al., 2009), but would probably reduce yield potential of corn and black oats as evidenced in our results. The defi nition of a critical limit is important to carry out the soil physical characteristics diagnosis and decisionmaking in soil or crop management.
In addition to the decrease in the macroporosity, we found an increase in microporosity, without changing the total pore volume in both evaluations. For this study with very clayey soil, the microporosity greater than the range from 0.46 to 0.51 m³ m -³ decreased the corn yield. The machine traffi c in our study caused an imbalance between macro and microporosity, which possibly affected hydraulic conductivity, despite maintaining total soil porosity. Hydraulic conductivity in soil is dynamic and usually expressed in terms of volumetric moisture without considering the pore size, as macro and micro porosity continuously varies in the soil without abrupt transition (REICHARDT; TIMM, 2020). There is possibly a plateau of micropores, and when soil compaction overcomes this limit at the expense of the macropore volume, it can increase water retained in the soil, or pores became unavailable for water retention, which in turn reduces the water availability and aeration to the roots. This is a hypothesis to explain the reduction in the corn and black oat yields in these aforementioned conditions; however, further studies with water availability approaches should be carried out to test this.
An increase in microporosity was caused by the soil compaction by the machine traffic treatments, which consequently decreased the macroporosity. This relationship of decreased macroporosity and increased microporosity agree with other studies (MARINS et al., 2018;MORAES et al., 2020), but most of them had no microporosity reference value.

CONCLUSIONS
This study has advanced in knowledge by identifying critical porosity, bulk density and degree of compactness values in a Xanthic Hapludox for developing corn and black oat crops under a no-till system. The 10% higher yields under the conditions of this experiment were obtained with macroporosity above 0.10 m 3 m -3 and 0.13 m 3 m -3 for corn and black oats, respectively. Similarly, the soil degrees of compactness of 90% and 86% were considered critical for the corn and black oat, respectively. In addition, soil compaction by machine traffic reduced maize yield up to 33% and black oat biomass yield up to 44%. Our results demonstrate that growers should be aware of the consequences of machine traffic on soil properties, as it can severely compromise crop yields, especially of corn and black oats cultivated in very clayey soil like the one evaluated in this study.