Introduction
No-till crop production in Brazil occupies more than 35 million hectares and continues to expand (FEBRAPDP 2013). About 85% of the total soybean [Glycine max (L.) Merr.] production area is no-till (EMATER 2014). The herbicide sulfentrazone, an inhibitor of protoporphyrinogen oxidase (PPO), is often used with glyphosate for preplant vegetation burndown in no-till soybean. In some cases, the use of glyphosate with PROTOX-inhibiting herbicides is synergistic, manifested in faster desiccation, reducing the time between preplant burndown and sowing (Jaremtchuk et al. Reference Jaremtchuck, Constantin, Oliveira Júnior, Biffe, Alonso and Arantes2008). The addition of a PROTOX inhibitor helps manage glyphosate-resistant weeds (Ashigh and Hall Reference Ashigh and Hall2010; Shaner Reference Shaner2000), and its residual activity allows for application of POST herbicides later in the season (Jaremtchuk et al. Reference Jaremtchuck, Constantin, Oliveira Júnior, Biffe, Alonso and Arantes2008).
Sulfentrazone belongs to the chemical family of triazolinones and the PROTOX inhibitor group. It is a weak acid with a dissociation constant (pK a ) of 6.56 (Grey et al. Reference Grey, Walker and Hancock1997); water solubility of 110 mg L−1 (pH 6.0); vapor pressure of 1.07 × 10−7 Pa (25 C) (Shaner Reference Shaner2014); partition coefficient (K ow ) of 9.8 (pH 7.0); and a half-life of 110 to 280 d in soil depending on the local edaphoclimatic conditions. It is degraded primarily by soil microbes (FMC Corporation 1995; Grey et al. Reference Grey, Walker and Hancock1997; Hess Reference Hess and Altman1993; Martinez et al. Reference Martinez, Silva, Fay, Maia, Abakerli and Durrant2010; Shaner Reference Shaner2012; Tomlin Reference Tomlin1994). Therefore, factors that favor high microbial population and activity also favor rapid degradation of sulfentrazone.
Despite broad adoption of mixing residual herbicides with nonselective herbicides for preplant burndown and application of PRE herbicides with glyphosate or following glyphosate application at planting, little is known about the dynamics of residual herbicides when applied to soil covered with plant residue. Recent studies indicate that residual herbicides can be effectively integrated in conjunction with cover crop termination applications in preplant, but termination timing and biomass accumulation will affect the amount of sulfentrazone reaching the soil (Whalen et al. Reference Whalen, Shergill, Kinne, Bish and Bradley2020).
The presence of plant residue on the soil surface can alter herbicide behavior (Selim et al. Reference Selim, Zhou and Zhu2003) and reduce herbicide efficacy. Plant residue prevents deposition of herbicide to soil. Herbicide molecules retained on plant residue are exposed to various avenues of dissipation, including volatilization and degradation by sunlight (Johnson et al. Reference Johnson, Wyse and Lueschen1989; Locke and Bryson Reference Locke and Bryson1997). The magnitude of these effects on herbicide molecules retained on plant residues depends on the occurrence of rain, which washes the herbicide into the soil, where it follows normal chemical dynamics in the soil environment (Banks and Robinson Reference Banks and Robinson1986; Ghadiri et al. Reference Ghadiri, Shea and Wicks1984; Reddy et al. Reference Reddy, Locke, Wagner, Zablotowicz, Gaston and Smeda1995).
In sugarcane (Saccharum officnarum L.), it was observed that more than 20% of the total sulfentrazone applied is retained in sugarcane straw (Carbonari et al. Reference Carbonari, Gomes, Trindade, Silva and Velini2016). There is practically no information about the dynamics of sulfentrazone applied in no-till soybean, whether on crop stubble or on cover plants at the time of desiccation. The prevailing hypothesis is that applying herbicide on soil with plant residue cover results in reduced herbicide efficacy.
The objectives were to evaluate the behavior and spatial distribution of sulfentrazone in the soil when applied on pearl millet [Pennisetum glaucum (L.) R. Br.], sorghum [Sorghum bicolor (L.) Moench ssp. bicolor], or bare soil.
Materials and Methods
Experimental Area
Two experiments were performed between 2013 and 2014 at the Lageado Experimental Farm, São Paulo State University–UNESP, Botucatu Campus, Sao Paulo, Brazil. The forage sorghum area was located at 22.844°S, 48.425°W, and the millet area at 22.844°S, 48.424°W, at an altitude of 779 m. The soil of the experimental areas was Dystrophic Red Nitosol (EMBRAPA 2006). The soil physical and chemical characteristics are listed in Tables 1 and 2. The climatic conditions during the implementation of the experiments are shown in Figure 1.
Chemical characterization of the soil sorghum and pearl millet cover crop locations.a
a Abbreviations: OM, organic matter; SB, sum of bases; V%, base saturation.
b There are several methods to determine phosphorus content in soil. In the extraction procedure with the resin, P moves into solution and is then adsorbed by the resin, a porous synthetic material with positive charges, mimicking what happens with roots.
Physical characterization of the soil in the sorghum and pearl millet cover crop locations.
Precipitation, irrigation, and accumulation (mm) during the experiments. 1st: collection performed 1 d after the application of sulfentrazone; 2nd: collection performed 31 d after the application of sulfentrazone; 3rd: collection performed 32 d after the application of sulfentrazone; 4th: collection performed 53 d after the application of sulfentrazone.
Experimental Design and Cover Crop Establishment
Forage sorghum and pearl millet were chosen due their common use as cover crops in Brazilian Cerrado. The forage sorghum cover crop was planted at 9 kg ha−1 and the pearl millet cover crop at 15 kg ha−1 of seed, both in a no-till system. The cover crops were sown on October 7, 2013. The experimental design was a randomized complete block design with four treatments and four replications. Each plot was 7-m long and 2.5-m wide, with five crop rows spaced 0.5 m apart.
Three preplant burndown treatments were tested on each cover crop: (1) glyphosate (Roundup® Original, 480 g ai L−1, Monsanto, km 159 Carlos Marcondes Road, Sao Jose dos Campos, Sao Paulo, Brazil) (720 g ae ha−1) at 20 d before sowing (DBS) followed by sulfentrazone (Boral 500 SC®, FMC, 943 Bortolo Jose Ferro Street, Paulinia, Sao Paulo, Brazil) (600 g ai ha−1) at 10 DBS; (2) glyphosate (720 g ae ha−1) mixed with sulfentrazone (600 g ai ha−1) at 10 DBS; and (3) sulfentrazone (600 g ai ha−1) at 10 DBS on bare soil. All treatments were performed in an area with a no-till planting system. For the treatment without cover crop, the cover crop plants were removed from the plots with minimal soil disturbance by using a hoe before the spraying herbicides.
Application of Treatments, Soybean Sowing, and Data Collection
At 1 d before herbicide application (20 DBS), cover crop residue samples were collected from four 0.25-m2 random sites per plot. No further biomass collection was made until all the termination treatments were applied and the cover crops were dead. At 9 d after the sulfentrazone application (DAA), 0.25-m² cover crop residue samples were collected per plot. The samples were stored in paper bags and dried in a forced-air circulation oven (60 C) until the samples reached constant mass. Total dry weights per plot were collected for data analysis.
The herbicides were applied using a CO2-pressurized backpack sprayer operating at a pressure of 300 kPa and a spray volume of 200 L ha−1. The sprayer was fitted with six flat-fan XR 11002 VS nozzles (TeeJet® Technologies, Spraying Systems, Wheaton, IL, USA) spaced 0.5 m apart.
Soybean ‘BMX Potência’ was planted in a no-till system on December 17, 2013, at 22 seeds m−1, on rows spaced 0.45 m apart (Strieder et al. Reference Strieder, Pires and Constamilan2015). Soybean was managed following region-specific recommendations. The crop was sprinkler irrigated as needed.
To determine the concentration of sulfentrazone in the soil profile, soil samples were collected at three sites per plot to create a composite sample per collection time (1, 31, and 53 DAA). The first collection was performed only in the 0 to 10 cm layer, before the occurrence of rain. The samplings at 31 and 53 DAA of sulfentrazone occurred after 90 mm of precipitation had fallen between the first and second sampling and between the second and third sampling, with a total of 180 mm of precipitation between 1 and 53 DAA (Figure 1). At these timings, the samples were acquired at three depths: 0 to 10 cm, 10 to 20 cm, and 20 to 40 cm. An auger-type probe was used to collect the soil samples, and the samples were placed in plastic bags and stored at −20 C until processing.
The purpose of any herbicide application is to distribute the product as evenly as possible in the treated area. To evaluate this distribution, in addition to the collections to determine the dynamics of sulfentrazone release in the soil profile, a spatial collection was also carried out, aiming to demonstrate the uniformity of the concentration of the herbicide at different points within the experimental plot in the different pre-sowing management systems for soybeans. The soil sampling was conducted at 32 DAA of the herbicide by collecting samples at five points from each parcel, which yielded a total of 20 samples per treatment at a depth of 0 to 10 cm, and the samples were stored as described earlier.
To extract the available sulfentrazone from the soil solution, the samples were thawed and dried at 40 C. After the samples were homogenized, a 7-g subsample of dry soil was removed and placed in a 10-ml plastic cartridge containing porous polyethylene. Each soil-containing cartridge was saturated with 2.5-ml of deionized water and refrigerated (8 ± 3 C) for 24 h. Water was used to remove only the available sulfentrazone fraction in the soil solution. The cartridges were then centrifuged (Hettich Zentrifugen centrifuge, Föhrenstraße 12 78532, Tuttlingen, Germany) at 3,270 × g and 25 C for 10 min (Carbonari Reference Carbonari2009). The soil-water extract was filtered using 3.0-ml plastic syringes with Millex® membrane filters (Millipore PVDF 13 mm × 0.45 µm, Merck Millipore, Tullagreen, Carrigtohill, County Cork, Ireland). The filtrate was transferred to 2.0-ml glass vials.
Sulfentrazone in the solution was quantified using high-performance liquid chromatography–mass spectroscopy with an HPLC column (Prominence UFLC, Shimadzu, Kyoto, Japan) coupled to a triple quadrupole mass spectrometer (3200 Q TRAP, Applied Biosystems, Foster City, CA, USA). A 50 × 4.60 mm C18 column with a 20-μl injection volume was used (Synergi 2.5-μ Hydro-RP 100 Å, Phenomenex, Torrance, CA, USA). The mobile phases used were 0.5% acetic acid in water (Phase A) and 0.5% acetic acid (Avantor Performance Materials, Center Valley, PA) in methanol (Avantor Performance Materials S.A., Xalostoc, Mexico) (Phase B), with a flow rate of 0.4 ml min−1. The following sequence of mobile phase gradients was used: (1) 70% Phase A and 30% Phase B for 1 min; (2) 5% Phase A and 95% Phase B, 3 min; (3) 5% Phase A and 95% Phase B, 2 min; (4) 70% Phase A and 30% Phase B, 2 min; and (5) 70% Phase A and 30% Phase B, 2 min. The mass spectrometer was operated in the positive electrospray ionization mode. The sulfentrazone multiple reaction monitoring was optimized to 386.9, and confirmation was achieved through two-ion transition (110.2 and 146.1). A standard curve of serial dilutions of sulfentrazone analytical standard with a certified purity level of 99.8% (supplied by FMC) was used for quantification.
Data Analysis
For the cover crop dry mass data, 95% confidence intervals were established. The data for sulfentrazone concentration in the soil were subjected to ANOVA by the F-test at 5% probability, and the treatment across depths means were compared using Tukey’s test (P ≤ 0.05). The data for the accumulated frequencies of the point availability of the herbicides were subjected to regression analysis using the Gompertz model in SAS (SAS Institute, Cary, NC) statistical software (Velini Reference Velini1995):
$$F = [{e^{a - {e^{\left( { - bc{\rm{*}}x} \right)}}}}]$$
where F represents the accumulated frequency of the availability of the herbicide as a percentage; a is the maximum asymptote of the curve, which was set to 4.60517; eª = 100; the displacement of the curve along the x axis is represented by the modulus of the parameter b; and the parameter c is the slope or concavity of the curve relative to the accumulated frequency.
Results and Discussion
The Effect of Glyphosate and Sulfentrazone Preplant Burndown on Cover Crop Residue
The cover crop termination treatments affected the amount of residue present at soybean planting for both sorghum and pearl millet cover crops and also the amount of residue in the soil surface by the time of sulfentrazone application. The sequential application of glyphosate (20 DBS) followed by sulfentrazone (10 DBS) reduced the amount of sorghum residue on the soil surface to 1,999 kg ha−1 compared with 9,480 kg ha−1 obtained with the tank mixture of glyphosate plus sulfentrazone by 9 d after sulfentrazone application and resulted in a better kill of the sorghum cover crop (Figure 2). Less plant residue on the soil surface has several consequences, including greater deposition of sulfentrazone on the top layer of the soil, which can led to better weed control. Sulfentrazone has foliar activity in addition to being active in the soil; therefore any living plants would be controlled by its application. In addition to that, in the tank-mixture treatment, the cover crops had 10 additional growing days, which would help account for the greater biomass in this treatment.
Average quantity of sorghum cover in the experimental areas at 70 d after sowing and at 9 d after the application of sulfentrazone. The sizes of the error bars are equal to the confidence interval (P ≤ 0.05). DBS, days before sowing; gly 20 DBS + sulf 10 DBS = glyphosate 20 DBS + sulfentrazone 10 DBS; gly + sulf 10 DBS = glyphosate + sulfentrazone 10 DAS.
As was observed with sorghum, the sequential application also resulted in less pearl millet straw on the soil compared with the tank mixture. In the pearl millet cover crop experiment, the amount of straw on the soil surface was 1,650 kg ha−1 in the sequential application of glyphosate and sulfentrazone and 5,090 kg ha−1 in the tank mixture (Figure 3). The presence of straw on the soil surface can affect the leaching of herbicides (Banks and Robinson Reference Banks and Robinson1982). Large amounts of herbicide may be retained in the straw, and rain or irrigation is needed to leach the herbicide from the straw to the soil to be effective on weeds. Correia et al. (Reference Correia, Camilo and Santos2013) determined that 20 mm of simulated rainfall after sulfentrazone application on the straw was sufficient to ensure that the herbicide reached the soil and attained a good level of weed control.
Average quantity of pearl millet cover in the experimental areas at 70 d after sowing and at 9 d after the application of sulfentrazone. The sizes of the error bars are equal to the confidence interval (P ≤ 0.05). DBS, days before sowing; gly 20 DBS + sulf 10 DBS, glyphosate 20 DBS + sulfentrazone 10 DBS; gly + sulf 10 DBS, glyphosate + sulfentrazone 10 DAS.
Behavior of Sulfentrazone in No-Till Soybean with Sorghum Cover Crop
Plots treated with glyphosate followed by sulfentrazone had a greater concentration of sulfentrazone in the soil solution than those treated with a tank mixture of both herbicides, but less than the amount of sulfentrazone detected in the soil of plots with bare soil at 1 DAA (Table 3). Thus, it seems a greater amount of cover crop residue may reduce the deposition of sulfentrazone into the soil and, therefore, intercepted more of the applied herbicide (Table 3). Similar results were reported by Whalen et al. (Reference Whalen, Shergill, Kinne, Bish and Bradley2020), who found that biomass accumulation had an inverse relationship with sulfentrazone soil concentrations, with lower sulfentrazone concentration occurring where cover crop biomass was greatest.
Concentrations of sulfentrazone in the soil at 1, 31, and 53 d after application (DAA) in the sorghum cover crop area.a
a DBS, days before sowing. Averages followed by the same lowercase letter do not differ from each other for the treatments according to Tukey’s test (P ≤ 0.05). Averages followed by the same uppercase letter do not differ from each other for the depth according to the Tukey’s test (P ≤ 0.05). NS, nonsignificant (P ≤ 0.05). *Significant (P ≤ 0.05). **Significant (P ≤ 0.01).
At 31 DAA, sulfentrazone concentrations in the top soil layer soil solution (0 to 10 cm) were reduced compared with 1 DAA with the sequential application and bare-soil treatments. In contrast, the tank mixture resulted in increased concentration of sulfentrazone from 1 DAA to 31 DAA in the same soil layer. This demonstrated that the greater amount of sulfentrazone that was retained in the high cover crop biomass was then leached into the soil profile after a 1-mo period plus 90 mm of precipitation. The same pattern was observed at 53 DAA in the top soil layer (0 to 10 cm) when compared with the levels of sulfentrazone observed at 1 DAA, where sulfentrazone concentration in the soil solution decreased in the sequential application and bare soil after a 1-mo period plus 90 mm of precipitation. Meanwhile, the sulfentrazone concentration increased from 10.79 ng g−1 (1 DAA) to 24.78 ng g−1 (31 DAA) when glyphosate and sulfentrazone were applied in mixture (Table 3). Rainfall and irrigation play a key role in releasing the herbicide retained in the straw to the soil, which would improve weed control (Carbonari et al. Reference Carbonari, Gomes, Trindade, Silva and Velini2016).
At 53 DAA, the concentration of sulfentrazone in the soil solution within the 10- to 20-cm depth decreased (for all treatments when compared with the concentrations observed at 31 DAA Table 3). At the 20- to 40-cm depth, the concentration in the soil solution increased with the sequential application and decreased in the tank mixture and bare-soil treatments (Table 3). The lower herbicide deposition into soil with a cover crop can reduce weed control. On the other hand, the cover crop residue itself can reduce the germination, emergence, and growth of weeds through physical impediment and/or allelopathy (Egley and Duke Reference Egley and Duke1985; Gomes and Christoffoleti Reference Gomes and Christoffoleti2008). This may make up for any loss in herbicide activity and allow for the use of cover crops for soil conservation and supplemental weed control.
With sugarcane residue, it takes 20 mm of rain to wash off sulfentrazone from the crop biomass covering the soil (Carbonari et al. Reference Carbonari, Gomes, Trindade, Silva and Velini2016). Therefore, more than enough rain or irrigation occurred in the current experiments to warrant the maximum release of sulfentrazone from the cover crop residues.
The main points of this study are that crop residue intercepts the herbicide, and the intercepted herbicide, with sufficient rain or irrigation, is leached into the soil within 2 mo. There is no evidence that soil moisture affects sulfentrazone degradation rate, although degradation happens faster at higher temperatures (Martinez et al. Reference Martinez, Silva, Fay, Maia, Abakerli and Durrant2008) decreasing the time in which the herbicide is effective on weed control. Microbial degradation also plays an important role in sulfentrazone dissipation in the soil (Ohmes et al. Reference Ohmes, Hayes and Mueller2000). The interception of herbicides by plant residues increases herbicide vulnerability to volatilization and photodecomposition before reaching the soil (Locke and Bryson Reference Locke and Bryson1997). This study also showed, indirectly, that sulfentrazone degradation by UV (while it is on the cover crop residue) is limited, because sulfentrazone concentration in the top soil layer increased with time after application (Shaner Reference Shaner2014). In other words, the largest proportion of herbicide molecule remained intact while on the residue, to be detected in larger concentrations with time in the soil. After the maximum leaching of sulfentrazone from cover crop residue into the soil had occurred, the difference in sulfentrazone concentration in soil solution no longer existed at any depth.
Regarding the spatial distribution of the herbicide in the plots (Figure 4; Table 4), the average availability of sulfentrazone in the soil solution was 21.52 ng g−1 for sequential applications, 17.57 ng g−1 with the tank mixture, and 42.63 ng g− 1 in the absence of crop straw. The distribution of sulfentrazone concentration data in soil without cover crop showed less dispersion and less variability than the soils with cover crops according to the adjustments provided by the Gompertz model. The tank mixture of glyphosate and sulfentrazone, which was sprayed in the presence of the highest amount of biomass, also had the highest dispersion of sulfentrazone concentration data and variability. This result indicates that in addition to interfering with the average sulfentrazone concentrations at different soil depths, the cover crop and manner of burndown reduced the uniformity of spatial soil distribution of sulfentrazone. Not only was the availability of sulfentrazone in the top soil layer lower with cover crop residue, its dispersion around the mean across the soil profile was even greater relative to the bare soil. This variability in soil herbicide concentration in the presence of cover crop residue is another important point to consider in weed management using cover crops, as it would result in increased patchiness of weed distribution, which may be addressed with variable-rate herbicide application technology in large-scale farms when that becomes commercially viable.
Accumulated frequencies of the point distributions of the sulfentrazone concentrations in the soils in the areas with sorghum cover. DBS, days before sowing.
Parameters from the Gompertz model and the averages, medians, and coefficients of variation (CV) from the data for the point distributions of sulfentrazone concentrations for the different modalities of application at the sorghum cover crop location.a,b
a Abbreviations: G, glyphosate; S, sulfentrazone; DBS, days before sowing.
b Gompertz model used: f = a*exp{−exp[−(x − c)/b]}. a = maximum asymptote of the curve; b = displacement of the curve along the x axis; c = slope or concavity of the curve relative to the accumulated frequency. **Significant by the F-test (P ≤ 0.01).
Behavior of Sulfentrazone in No-Till Soybean with Pearl Millet Cover Crop
At 1 DAA, the sulfentrazone concentrations in the soil solution with the sequential application and in the bare-soil treatments were similar, whereas less herbicide was found with the tank-mixture application (Table 5). The amount of sulfentrazone in the soil solution that reached the top soil layer with pearl millet as cover crop was 52% greater with sequential application and 135% greater with the tank mixture when compared with the area with sorghum as the cover crop.
Concentrations of sulfentrazone in the soil at 1, 31, and 53 d after application (DAA) in the pearl millet cover crop area.
a DBS, days before sowing. Averages followed by the same lowercase letter do not differ from each other for the treatments according to Tukey’s test (P ≤ 0.05). Averages followed by the same uppercase letter do not differ from each other for the depth according to the Tukey’s test (P ≤ 0.05). NS, nonsignificant (P ≤ 0.05). *Significant (P ≤ 0.05). **Significant (P ≤ 0.01).
Sulfentrazone concentration in the soil solution decreased between soil collections for all treatments and all depths, reducing the differences among treatments throughout the time of the study. At 31 DAA, the sulfentrazone concentrations in the top 0 to 10 cm of soil were 37%, 16%, and 61% lower when compared with the concentrations in the first sampling, respectively, at the same depth. Similar dynamics occurred at 53 DAA, when the sulfentrazone concentrations declined by 75%, 27%, and 78% for the same treatments at 0 to 10 cm (Table 5).
At the 10- to 20-cm depth, the decline in sulfentrazone concentration between 31 and 53 DAA was 69% in the treatment with sequential applications, 53% with the tank mixture, and 47% in bare soil (Table 5). The decline in sulfentrazone between 31 and 53 DAA was also apparent at the 20- to 40-cm layer, with 72%, 25%, and 16% for sequential application, tank mixture, and without pearl millet cover crop, respectively (Table 5).
Using the Gompertz model, cumulative frequency curves of sulfentrazone concentration in the soil solution at different sampling times and depths were developed (Figure 5; Table 6). The average sulfentrazone concentration in the soil solution was 68.43 ng g−1 with sequential applications, 38.38 ng g−1 using the tank mixture, and 56. 91 ng g−1 in the absence of pearl millet, which again shows reduction in the availability of herbicide in the soil due to the presence of the cover crop, especiallyif plants are alive. Sulfentrazone application on bare soil and the application on partially dead cover (sequential to glyphosate) resulted in greater dispersion of data around the mean. In contrast to the sorghum cover crop experiment, the concentration of sulfentrazone in soil with pearl millet killed by sequential application of glyphosate and sulfentrazone was greater than in the bare soil. As in the sorghum experiment, the tank mixture of glyphosate and sulfentrazone on pearl millet cover crop at 10 DBS caused the greatest variation in the spatial soil distribution of sulfentrazone and the lowest average concentration of the herbicide.
Accumulated frequencies of the point distributions of the sulfentrazone concentrations in the soils in the areas with pearl millet cover. DBS, days before sowing.
Gompertz model parameters and the averages, medians, and coefficients of variation (CV) for the data from the point distributions of the sulfentrazone concentrations for the different modalities of application at the pearl millet cover crop location.a,b
a Abbreviations: G, glyphosate; S, sulfentrazone; DBS, days before sowing.
b Gompertz model used: f = a*exp{−exp[−(x − c)/b]}. a = maximum asymptote of the curve; b = displacement of the curve along the x axis; c = slope or concavity of the curve relative to the accumulated frequency. **Significant by the F-test (P ≤ 0.01).
Lessons Learned from the Two Cover Crop Experiments
Understanding how the cover crop alters the dynamics of sulfentrazone in the soil can support the adjustment of herbicide doses and management recommendations, ensuring that the quantity of herbicide in the soil is sufficient and herbicide distribution is uniform to achieve optimum weed control.
Less cover crop residue is expected to result in better herbicide activity than high cover crop residue. Less residue also facilitates planting of the main crop (i.e., soybean) and is expected to result in better crop establishment, although the yield effect is uncertain and depends on each crop–soil ecosystem (Forcella et al. Reference Forcella, Buhler, McGiffen, Hatfield and Stewart1994). Studies similar to the ones we are reporting here are needed to determine the optimum combination of cover crop residue and herbicide dose to achieve the best possible level of crop establishment, weed control, and soil conservation. The straw from the cover crops may improve weed control, but if the amount of mulch is competitive enough to suppress weeds, it may also be enough to interfere with crop growth and yield (Teasdale Reference Teasdale1996). As reported by Whalen et al. (Reference Whalen, Shergill, Kinne, Bish and Bradley2020), growers can integrate residual herbicides with cover crops to eliminate an additional sprayer pass; however, applying the residual herbicide preplant in conjunction with the termination application provided higher overall weed control than an early residual herbicide application.
The herbicide behavior dynamic was the same with both pearl millet and sorghum cover crops. The initial sulfentrazone concentration in the soil solution in the area with sorghum cover was lower than in the area with pearl millet at the 0- to 10-cm depth, because sorghum biomass was greater than that of pearl millet. Regardless of cover crop species, the plant residue retained a large proportion of the herbicide. The sulfentrazone leached deeper into the soil profile and its concentration diminished with time. In both studies, a portion of the herbicide eventually leached to the 20- to 40-cm layer, where it was not needed.
Cover crops affected the dynamics of spatial distribution and dissipation of herbicide in soil. The amount of sulfentrazone herbicide intercepted by the cover crop residue increased with the amount of biomass. Live cover crop intercepted more herbicide than partially dead cover crop. Live cover crop caused the highest variability in horizontal and vertical distribution of sulfentrazone in soil, and this variability increased with the amount of biomass.
Managing cover crops with a sequential application of glyphosate followed by sulfentrazone is better than with a tank mixture of both herbicides, because partial desiccation with glyphosate allows more sulfentrazone to be deposited directly on the soil.
Acknowledgments
The authors thank the Brazilian Federal Agency for the Support and Evaluation of Graduate Education (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior–CAPES) for their financial assistance. No conflicts of interest have been declared.
