No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado

ABSTRACT No-till planting and the use of organomineral fertilizers are crop management practices that increase soil organic matter, thereby mitigating leaching and cycling a considerable amount of nutrients, with the potential to improve crop yield. This study aimed to assess the agronomic performance of cabbage grown under a no-till system, using different cover crops and phosphorus sources. A randomized block design was used, with the main plot consisting of eight split plots for different cover crop treatments: 1) Signal grass (SG); 2) Sunn hemp (SH); 3) Pearl millet (PM); 4) SG+SH; 5) SG+PM; 6) SH+PM; 7) SG+SH+PM; 8) conventional tillage (soil preparation with no cover crop), and phosphorus (P) sources in the sub-plots: 1) mineral fertilizer (FM); 2) organomineral fertilizer (OF); 3) no P, with four repetitions. The following characteristics were assessed: cover crop fresh (FW) and dry weight (DW) (t/ha), residue decomposition and nutrient cycling; and cabbage head FW and DW (HFW and HDW) (g/plant) and yield (YLD) (t/ha). The highest FW and DW were recorded in the intercropped cover plant treatments; PM+SH and SG+SH residue exhibited the highest decomposition rate and P cycling into the soil. The highest cabbage HFW and YLD occurred in the SG+SH treatment, regardless of the fertilizer used. The MF used as P source produced a greater cabbage YLD when grown in PM residue. Under conventional tillage, YLD was higher when OF was used as P source.


Received on October 18, 2022; accepted on April 11, 2023
Brassicas are among the most widely consumed vegetables in Brazil due to its high nutrient content, commercial value and ability to grow in any region of the country, cabbage being one of the most irrigation, which can increase water erosion and nutrient leaching (Torres et al., 2021).This management system compromises soil attributes, causing yield loss (Santos et al., 2017) V egetable farming in Brazil is almost always associated with the intensive use of agricultural equipment for tillage, large amounts of soluble mineral fertilizers and sprinkler No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado important, as a quality food with high levels of potassium, calcium, fibers, sugars, citric and ascorbic acid, and vitamins A and C (Torres et al., 2015(Torres et al., , 2017;;Vieira et al., 2020).
No-till vegetable production (NTVP) is a technological innovation used to replace conventional tillage and has maintained or increased crop yields (Massan et al., 2019).A more recent technology is organomineral fertilizers (OFs), a mixture or organic matter and mineral fertilizer (Vieira et al., 2020).
OF application has improved the yields of crops grown under conventional tillage because organic matter (OM) in the fertilizer contributes to increasing the cation exchange capacity of the soil, reducing nutrient loss by leaching (Silva et al., 2020).This organic matter slowly releases nutrients that supply the needs of the crops throughout their growth cycle (Rodrigues et al., 2016).
The presence of a larger amount of organic anions in these OF blends results in greater competition for phosphorus (P) adsorption sites, thereby increasing plant-mineral interaction, reducing P fixation and decreasing P 2 O 5 transformation into forms unavailable to plants (Sousa et al., 2013).
Although some studies with OFs applied to vegetables grown under conventional tillage have reported good yields in some crops, little is known about their use in areas under NTVP, where organic matter is continuously supplied after each crop cycle (Torres et al., 2019(Torres et al., , 2021;;Silveira et al., 2021).
Fertilization with OFs in vegetable growing areas may provide new insights on production and improve the physical and chemical quality of soil.However, since vegetables have a short growth cycle, the use of these slow-release fertilizers requires further investigation to safeguard against nutrient deficiencies in these crops.As such, this study aimed to assess the agronomic performance of cabbage grown under a no-till system in Uberaba, Minas Gerais state (MG), Brazil, using different cover crops and phosphorus sources.

MATERIAL AND METHODS
The study was conducted from November 2020 to July 2021, in an experimental area in its first crop cycle, on the Uberaba Campus of the Federal Institute of Education, Science and Technology of the Mineiro Triangle (IFTM) (19°39'19"S; 47º57'27''W, 800 m altitude), using an NTVP system.
No pH correction was performed due to insufficient liming time.
Climate in the region is Aw, classified as warm tropical according to the updated Köppen classification (Beck et al., 2018), with hot rainy summers and cold dry winters.Average annual rainfall, temperature and relative humidity are 1,600 mm, 22.6ºC and 68%, respectively (Inmet, 2021); however, 791 mm and 24°C were recorded during the study period (Figure 1).
With respect to the P sources studied, granulated MAP contained 11% N and 52% P 2 O 5 ; and the pelleted OF supplied by Vitória Fertilizantes S.A. contained 6% N and 30% P 2 O 5 , with only N corrected according to the concentration of each source.
The area was used for forage (Urochloa brizantha cv marandu) over an extended period of time, followed by two successive soybean and corn crops, and then left fallow for a year before planting cabbage under an NTVP system.
In the no-till system used here, pearl millet was chosen as the first cover crop to provide mulch across the entire area, followed by the other cover plants.The cover crops were planted in experimental units measuring 5 m long by 6 m wide (30 m²).
Planting was carried out using a Semina 2 seed drill, with drill rows spaced 0.20 m apart, using 50, 50 and 25 signal grass (SG), pearl millet (PM) and sunn hemp (SH) seeds, respectively, for the individual cover crop treatments.For intercropped SG+SH, SG + PM, PM+SH, 50% of each seed dose was used, and 33% for the triple intercropped treatment (SG+SH+PM).
The cover crops were grown from October 13, 2020 to January 25, 2021 with no mineral or organic fertilizers.Since cover crops flower at different times, when 50% of the plants in the entire area reached peak bloom, a 2 m 2 area in each experimental unit was sampled for fresh weight assessment.Next, the material was dried in an oven at 65ºC for 72 hours or until constant weight and then weighed to determine dry weight, with results expressed in t/ha.
At the end of the growth period the plants were desiccated using glyphosate (Gliz ® 480 SL, 960 g/ha ai) and 2.4-D dimethylamine (Aminol ® 806, 1612 g/ha ai) with a spray volume of 250 L/ha, and cut close to the ground with a backpack brush cutter.Next, holes were made in the cover crop residue for subsequent fertilization and transplanting of the cabbage (Brassica oleracea var.capitata) seedlings.
The seedlings were produced from cv.Astrus seeds with a 90-day cycle, in 128-cell polystyrene trays filled with commercial Bioplant substrate and placed in a greenhouse with plastic roofing and closed sides.
On March 25, 2021, at a height of 10 to 15 cm, the cabbage seedlings were transplanted to holes (approximately 1.15 m wide and 0.18 m deep) previously made in the cover crop residue using a sow dibbler, with 0.70 x 0.50 m spacing.
The sub-plots with phosphate fertilizer consisted of four 4-meter long rows containing four plants per row, totaling 16 plants per plot with corridors between the experimental blocks.The four center plants were considered the study area for analyses.
The fertilizer doses applied to the cabbage plants were established based on soil analysis and the recommendations of the Soil Fertility Committee of Minas Gerais State (Fontes et al., 1999).The recommended nitrogen (N), phosphorus (P) and potassium (K) doses for cabbage were 150 kg/ha of N, 400 kg/ha of P 2 O 5 and 240 kg/ha of K 2 O. N and K were applied at transplanting and 30 and 45 days after transplanting (DAT), using 50 kg/ha of N and 80 kg/ha of K per application, whereas the full P dose was manually applied to each hole one day before transplanting.
Foliar spraying was carried out at 15, 30 and 40 DAT to supply boron (B), molybdenum (Mo) and cobalt (Co), using a solution of 1 g/L of boric acid (17% B) and 2.7 mL of Vitaphol CoMo ® , which contains 10% Mo and 2% Co. Applications aimed to completely cover the leaves without allowing the spray solution to drip.
The plants were irrigated daily as needed, using a conventional sprinkler system consisting of spray heads spaced 9 m apart with a flow rate of 560 L/h and irrigation time of 20 min/day to maintain soil moisture content close to field capacity.
Weeds were controlled via a single application of fluazifop-p-butyl (Fusilade ® 250 EW, 175 g/ha of ai) at 10 DAT and by manual weeding throughout the crop cycle.
Litter bags were used to quantify the decomposition rate of cover crop residue, as described by Santos & Whilford (1981).The 0.04 m² bags (0.20 x 0.20 m) were made from 2 mm mesh nylon and contained 30 g of cover crop shoots, previously oven dried at 65ºC for 72 hours or until constant weight.
Immediately after cabbage seedling transplanting, four nylon litter bags were distributed in each experimental unit and collected 90 days later.The bags were sampled at distribution to determine the baseline (time zero) and identify the onset of decomposition.The bags were collected from the field after 90 days, their contents removed and then oven dried at 65ºC for 72 horas or until constant weight.Each sample was manually cleaned to remove any mineral residue and subsequently determine dry weight, expressed in t/ha.
The exponential mathematical model described by Thomas & Asakawa (1993) was applied to calculate plant residue decomposition using the equation X = Xoe -kt , where X is the dry weight remaining after time t, in days; Xo the initial amount of dry matter and k the decomposition constant.
Based on the k value, the half-life (T 1/2 ) of the remaining residue was calculated using the formula T 1/2 = 0.693/k, proposed by Paul & Clark (1996), which expresses the time taken for half the residue to decompose or for half of the nutrients in the residue to be released.
In order to quantify nutrient cycling in the cover crop residue, the material removed from the litter bags was ground and sent to the laboratory for chemical analysis.
Total N was determined by the Kjeldahl method, P by colorimetry and K by spectrophotometry (Teixeira et al., 2017).Ca and Mg were measured by atomic absorption spectrometry and S by turbidimetry (Tedesco et al., 1995).Extraction of each nutrient was estimated based on the percentage of nutrient present in each sample, multiplied by the total dry weight.
Cabbage was harvested when head compactness (firmness) reached commercial grade, at 83 DAT, over a period of 10 days.After harvesting, the heads were sent to the laboratory to determine head fresh (HFW) and dry weight (HDW) and yield (YLD).
Assumptions of normality and homogeneity of residual variances were verified using the Shapiro-Wilk and Bartlett tests, respectively.The values obtained for the characteristics studied were submitted to analysis of variance using Agroestat statistical software.The F test was applied and, when significant, the means for the cover crops were analyzed using the Scott-Knott method.Decomposition rates were submitted to regression analysis in SigmaPlot software version 12,5.Significance was set at 5% for all analyses.

RESULTS AND DISCUSSION
The SG+PM and SG+PM+SH cover crop treatments obtained the highest FW production, at 50.2 and 53.2 t/ ha, respectively, while the largest DW values were recorded for PM+SH (14.4 t/ha) and SG+PM+SH (15.7 t/ha).Sunn No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado hemp exhibited the lowest FW and DW when compared to the other cover crops, with 26.8 and 7.5 t/ha, respectively (Table 1).
The FW and DW values obtained can be considered high for the region, which can be justified by the rainfall during the study period (October 13, 2020 to January 25, 2021), with a total of 915.0 mm and average temperature of 24.0°C.
The higher respective FW and DW values for SG+PM and SG+PM+SH; and PM+SH and SG+PM+SH when compared to the individual cover crops (Table 1) confirm the benefits of intercropping these species, such as increased biomass production due to the balanced carbon-to-nitrogen ratio (C:N) achieved between the intercropped species and better soil exploration via different root systems.
Intercropping cover plants, especially grasses and legumes, reduces nitrogen immobilization by soil microorganisms, promoting an increase in soil nutrient content and dry matter accumulation and greater water and nutrient use efficiency due to better soil exploration by the different root systems (Latati et al., 2016).Under tropical conditions such as those in the Cerrado, this is an interesting alternative to ensure mulch formation and increase soil organic matter content (Rodrigues et al., 2012).Torres et al. (2017) and Mazetto Junior et al. (2019) studied FW and DW production in the same region and growing season and found FW values of 4.4 to 41.6 t/ha for SG, 8.5 to 41.7 t/ha for SH and 8.5 to 46.5 t/ ha for PM, whereas the intercropped treatments SG+SH, SG+PM, SH+PM and SG+PM+SH obtained values of 19.1 to 39.0, 17.8 to 37.1, 11.1 to 44.8 and 28.7 to 49.8 t/ha, respectively.The results for DW were 3.4 to 13.2 t/ha for SG, 5.0 to 9.8 t/ha for SH and 6.1 to 14.2 t/ha for PM, with respective values of 7.2 to 10.0, 6.9 to 13.2, 6.9 to 13.4 and 8.0 to 13.7 t/ha for SG+SH, SG+PM, SH+PM and SG+PM+SH.In both studies, rainfall distribution in the region was a decisive factor in cover crop FW and DW production, with values similar to those observed in our study.
These same cover crops (SG, PM and SH) and combinations were assessed by Torres et al. (2015;2017) and Silveira et al. (2021) in NTVP systems and showed promising results since, in addition to providing protection against soil erosion, they maintained soil moisture content for longer, reduced the volume of irrigation water needed and cycled a considerable amount of nutrients, thereby improving the agronomic indicators of the subsequent commercial crop.
Analysis of the decomposition rate of the different cover crops studied showed that, after 90 days, residue of 4.4 (58% of the initial total), 61 (63%) and 8.5 t/ha (75%) remained for SH, SG and PM grown alone and 6.8 (62% of the initial total), 10.3 (66%), 7.9 (71%) and 11.5 t/ha (73%) for SG+SH, SG+PM+SH, PM+SH and SG+PM, respectively (Figure 2).These findings indicate that decomposition was slower for PM grown alone or intercropped with other plants, which can be justified by the higher C:N between the plants used (Silveira et al., 2021;Torres et al., 2021).
On the other hand, SH exhibited the highest remaining DW after 90 days when grown individually or in No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado the intercropped treatments (Figure 2).Similar findings were obtained in other studies with this species when grown alone or intercropped with SG and PM (Soratto et al., 2012;Algeri et al., 2018).Greater decomposition of SH residue occurs because these plants generally have a low C:N ratio due to their high biological nitrogen fixation (BNF) (Soratto et al., 2012;Pacheco et al., 2013;Torres et al., 2017).Whether grown individually or intercropped, PM reduced the decomposition rate of residue, possibly because of its higher C:N ratio when compared to the other cover crops in the present study, as observed by Algeri et al. (2018), Mazetto Junior et al. (2019) and Torres et al. (2021) in research conducted in the Cerrado.
The half-life (T ½ ) indicates when 50% of the cover crop residue has decomposed and can be estimated using the constant (k) of the decomposition curve equations (Table 2).
SH and SG grown alone, together (SG+SH) and intercropped with PM (SG+SH+PM) obtained lower T½ values than those of the other treatments.This can be explained by the high BNF of SH and the fact that SG had not reached peak bloom at cutting because of its slow initial development when compared to the other cover crops studied.
According to Collier et al. (2018), using Poaceae and Fabaceae as individual or combined cover crops improves FW and DW production by increasing soil organic matter content, which improves soil quality, since the plants are adapted  * and ** = Means followed by the same lowercase letter in the rows and uppercase in the columns do not differ according to the Scott-Knott test (p<0.01and 0.05).SG = signal grass; PM = pearl millet; SH = sunn hemp; CV = coefficient of variation.
Horticultura Brasileira, v. 41, 2023 No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado to the soil and climate conditions in the Cerrado.In the present study, PM+SH and SG+PM+SH exhibited the highest DW production (Table 1, Figure 2), but PM alone obtained the highest T ½ and its intercropped treatments the lowest (Figure 2), demonstrating that PM residue directly affects the decomposition dynamics of soil cover, favoring an intermediate C:N in the intercropped treatments, especially those containing SH.The combination of SG and SH produced a lower T½ in relation to SG+PM and PM+SH (Figure 2), resulting in greater residue decomposition and nutrient cycling.The same behavior was observed by Mazetto Junior et al. (2019), which, when combined with the BNF of SH and subsequent N availability in the soil after cutting, results in less mobilization of the N supplied via mineral fertilizers by soil microorganisms (Ferreira et al., 2018), ensuring greater availability of the nutrient for the production system.
Greater N availability associated with high temperatures and adequate soil moisture content tends to accelerate residue composition, which can be up to three times faster when compared to cultivated areas under natural climate conditions (Torres et al., 2019).Silveira et al. (2021) reported T½ of 28 days for SG (k = 0.025), 80 days for PM (k = 0.087), 41 days for SH (k = 0.017), 43 days for SG+SH (k = 0.016), 69 days for SG+PM (k = 0.010), 77 days for PM+SH (k = 0.009) and 69 days for SG+SH+PM (k = 0.010).Except for SG, all these values are higher than those recorded in our investigation (Table 2), which is directly linked to the climate conditions in each study.By contrast, the T½ values of Torres et al. (2021) were lower than those obtained here, with 25 days for SG (k = 0.027), 28 days for PM (k = 0.024), 23 days for SH (k = 0.030) and 29 days for PM+SH (k = 0.024).
SG is a Poaceae that, when cut at the same inflorescence stage as other plants, generally has a similar C:N ratio compared to that of PM which, in turn, is typically higher than that of SH (Pacheco et al., 2013).However, in the present study, SG was cut at the onset of flowering, which explains the high residue decomposition rate (Figure 2) and its low T½ value (Table 2).Algeri et al. (2018) assessed the biomass production and soil cover of individually-grown and intercropped SG, PM and SH and found that, when grown alone, SG develops more slowly than the other plants, which take longer to cover the soil, and its residue contains more accumulated carbon, meaning that it tends to decompose faster than PM and at a similar rate to SH, behavior also observed in our study.
According to Collier et al. (2018) and Torres et al. (2021), using different cover plants and maintaining their residue on the soil before planting c o m m e r c i a l c r o p s c a n r e d u c e dependence on mineral fertilizer.This is because the mineralization of organic matter increases nutrient availability in the soil, largely due to the high BNF of legumes and the release of phosphate cations and anions into the soil solution, which occurs more rapidly in areas under NTVP, since high temperatures and moisture content contribute to increasing soil microbial activity, accelerating OM decomposition (Mazetto Junior et al., 2019).Pacheco et al. (2013) analyzed nutrient accumulation in SG and PM and obtained the following values: N (116.1 and 29.3 kg/ha), P (10.4 and 2.6 kg/ha), K (92.9 and 12.7 kg/ha), Ca (53.8 and 8.9 kg/ha) and Mg (11.6 and 3.8 kg/ha).With respect to P content in the cover crop residue, SH (12.5 g/ha) and PM+SH (12.0 g/ha) exhibited higher values at cutting when compared to the other cover plants, which also occurred at 90 days in the treatments without P (9.1 and 9.3 g/kg), with OF (7.4 and 8.1 g/kg) and MF (2.8 and 7.3 g/kg, not differing from SG at 3.5 g/kg), whereas SG+PM obtained the lowest P content at cutting and at 90 days (Table 3).
For accumulated P, the highest values were recorded for PM+SH at cutting (188.4 kg/ha) and 90 days after cutting, regardless of the P sources used (Table 3).The high values obtained in PM+SH without P (146.0 kg/ha) and with OF (127.2 kg/ha) and MF (114.6 kg/ha) are due to the greater DW production (15.7 t/ha) (Table 1) and remaining DW in this treatment (Figure 2), followed by SG and SH at cutting (91.1 and 93.8 kg/ ha) in treatments without P (24.5 and 68.3 kg/ha) and with OF (12.7 and 55.5 kg/ha) and MF (34.3 and 21.0 kg/ha), related to low DW production (9.8 and 7.5 t/ha, respectively) (Table 1).Collier et al. (2018) analyzed soil chemical attributes and the association between the residual effect of nutrient cycling in Fabaceae stubble and maize yield and concluded that sunn hemp is an advantageous option for total nutrient accumulation, since it increases available P levels in the soil and, consequently, the productivity of maize cobs, with values 24% higher than those recorded in maize grown on spontaneous vegetation.
Accumulated P tends to be released into the soil more rapidly for SG and SH and PM+SH, since these cover crops exhibited T½ of 29, 29 and 33 days, respectively (Table 2), values significantly lower than those recorded for PM (41 days) and SG+PM (39 days).
In areas under no-till production (NTP), the decomposition of Poaceae and Fabaceae used in crop rotation releases low-molecular-weight organic acids, which can block adsorption sites and thereby increase plant-available P, as reported by Maia et al. (2015).
The slower release of remaining P in dry matter is linked to the fact that, like most diesters, nucleic acids, phospholipids and phosphoproteins, P is less soluble in water because its release from plant tissue depends on soil microorganism activity.According to Khatounian (1999), P released from plant residue during mineralization can be absorbed by the subsequent commercial crop after the cover plants have been cut or be fixed in difficult-todissolve mineral compounds.
There was significant interaction between the cover crops and P sources for all the productive variables studied (Table 4).Unfolding of the significant interactions indicated that even when P was not applied, SG produced the lowest cabbage HFW (521.2 g), followed by PM+SH and the conventional treatment (Table 5).Similarly, when OF was used, the lowest yields were observed for cabbage grown in SG (1550.5 g) and under conventional tillage (1558.6 g), and with mineral fertilization, cabbage HFW was lower in SG, PM+SH, SG+PM+SH and the conventional treatment than for the remaining cover crops.The different P fertilizers had no effect on cabbage HFW in PM, SH and SG+SH (Table 5).
For all the cover crops assessed, OF resulted in higher cabbage HFW than that observed for MO or in treatments with no P source, except for SG+PM, which exhibited lower HFW values than those obtained with 100% mineral fertilizer.
According to Maia et al. (2015), available P is sensitive to variations in soil moisture content, which are common under NTP systems because soil moisture is maintained for longer, whereas conventional tillage promotes lower soil water levels with a resulting decline in available P, since diffusion depends on water.
For cabbage HDW, the highest values without P application were recorded in SG+PM (101.1 g/plant), with OF application in SH (107.5 g/ plant), SG+SH (113.0 g/plant), PM+SH (95.8 g/plant) and SG+SH+PM (99.3 g/ plant), and with MF in the PM treatment (109.6 g/plant).Comparison of the P sources demonstrated that cabbage HDW was highest in SG+SH+PM for all the sources, including the treatment with no P application (Table 5).
Generally, the lowest cabbage HFW, HDW and YLD values were always obtained with SG or under conventional tillage without soil cover, which confirms the importance of cover crop residue to mitigate problems such as erosion and leaching, and cycle considerable substantial amounts of macro and micronutrients into the soil, particularly N from biological fixation by sunn hemp.These findings are corroborated by previous research by Collier et al. (2018), Silveira et al. (2021) and Torres et al. (2021) under natural conditions and in irrigated areas.
However, the importance of interaction between the cover crop and P source was evident in the significant differences between HFW, HDW and YLD for the fertilizers used, whereby higher values were generally observed in areas with a cover crop when compared to the conventional treatment.
It should be noted that OF and MF application increased YLD in PM, SG, PM+SH, PM+SG+SH and conventional tillage when compared to no phosphate fertilization (Table 5).Among the cover crops studied, PM+SH exhibited greater remaining DW and accumulated P than the other treatments, regardless of the P source used.
However, PM+SH produced higher cabbage YLD values with OF and MF application when compared to treatments with no P.
The highest fresh and dry weight values were recorded when cover plants were intercropped.
Residue from the PM+SH, SH and SG treatments had the lowest T 1/2 , fastest decomposition rate and highest P cycling.
The highest HFW values were obtained in PM, SH and SG+SH without P application or with the use of MF or OF.
The largest cabbage yields were observed in SG+SH without P fertilization or with the use of MF or OF.
Mineral fertilizer used as P source produced the highest yield when cabbage plants were grown in PM.
Under conventional tillage, cabbage yields were larger when OF was applied as a P source.

Table 1 .
Fresh (FW) and dry weight (DW) production of different cover crops under a no-till cabbage growing system in the Brazilian Cerrado, in Uberaba-MG.Uberaba, IFTM, 2021.
** = Significant (p<0.05).Means followed by the same lowercase letter in the column do not differ according to the Scott-Knott test (p = 0.05).SG = signal grass; PM = pearl millet; SH = sunn hemp.

Table 4 .
Mean head fresh weight (HFW), dry weight (HDW) and yield (YLD) of cabbage grown under different cover crop residues and P doses on a no-till system in Uberaba-MG.Uberaba, IFTM, 2021.No-till cabbage production in different cover crops and phosphorus sources in the Brazilian Cerrado

Table 5 .
Unfolding of the interaction between cover crops and phosphate fertilizers for head fresh (HFW) and dry weight (HDW) production and yield (YLD) of cabbage grown under a no-till system in Uberaba-MG.Uberaba, IFTM, 2021.