Introduction

Tobacco plants are from the family Solanaceae, whose main commercial representative is the species Nicotiana tabacum L., which can synthesize nicotine (Silva-Vignoli & Mentz, 2005; Bailey, 2013). Brazil is the world´s largest exporter and the third-largest producer of tobacco; approximately 80 % of Brazilian tobacco production is intended for export (Food and Agriculture Organization of the United Nations [FAO], 2022). The South of Brazil is the main tobacco producing region in the country (Becker et al., 2020). This region was responsible for a production of 560,181 kg of leaf tobacco in 246,590 ha in the 2021/2022 harvest season, involving 128,448 growers (Associação dos Produtores de Fumo do Brasil [Afubra], 2022). Rio Grande do Sul is the main tobacco producing state in Brazil, responsible for 44.2 % of the national production, followed by Santa Catarina (30.7 %), and Parana (25.1 %) (Instituto Brasileiro de Geografia e Estatística [IBGE], 2022). Considering the total tobacco grown in the South region of Brazil, the mean yield was 2,272 kg∙ha^-1 in the 2021/2022 harvest (Afubra, 2022).

In Brazil, tobacco is produced mainly on small rural properties by family farmers, which perform all production phases manually (Vargas & Oliveira, 2012). The main tobacco varieties used are Virginia (91.5 %) and Burley (7.5 %) (SindiTabaco, 2022), and the crop is cultivated in sloping to undulating areas, with soil and climate favorable. These producing farms usually grown other species, such as grasses for pastures, maize, common bean, and vegetable crops, in addition to tobacco, which is grown in 2 to 4 ha due to its high profitability (Antoneli & Bednarz, 2011; Becker et al., 2020).

Tobacco has been produced in integrated production systems involving companies and growers. The development of technologies and research has increasingly improved production, resulting in increased tobacco quality and yield (Silveira, 2015). This improvement may be connected to methods and techniques of cultivation, such as use of the no-tillage system (NTS), which causes less soil mobilization (Antoneli & Thomaz, 2014).

Tobacco is usually cultivated on ridges to promote better root development (Antoneli et al., 2016). Under NTS, the plant cover is desiccated and its straw on the ground keeps the soil covered. Thus, weeds are controlled with the application of herbicides, avoiding soil turning by plowing or manual weeding (Antoneli & Thomaz, 2014). According to Ritter et al. (2005), the presence of weeds decreases tobacco yield by 25 %, as they directly affect leaf yield. In addition, weeds hold pathogens, insects and nematodes, and they also make cultural practices difficult. Bailey (2013) reported decreases between 28 to 40 % in total tobacco yield when compared to that found in plots treated with herbicides.

Despite the importance of herbicide application for tobacco production, only five active ingredients are currently registered for pre-emergence application on tobacco plantations for controlling weeds (Agrofit, 2023), including sulfentrazone and clomazone, which are the most used by tobacco growers (Clapp et al., 2022). However, no single herbicide can effectively control all the weeds that can occur in these plantations, thus requiring the application of combinations of different active ingredients (Bailey, 2013).

Sulfentrazone is a pre-emergent herbicide from the chemical group of isoxazolidinones and is recommended for application to tobacco plantation areas to control weed species of the Magnoliopsida and Liliopsida classes (Pekarek et al., 2010; Clapp et al., 2022). Clomazone is a herbicide from the chemical group of isoxazolidinones and inhibits the enzyme deoxy-D-xylulose-5-P synthase (DXS), causing bleaching in sensitive plants; it is recommended for pre- and post-emergence applications to tobacco plantation areas to control annual grass species (Bailey, 2013; Darwish et al., 2015).

Successful chemical weed control depends on the selectivity of herbicides (Clapp et al., 2022). According to Dan et al. (2011), this selectivity is based on the ability of the plant to metabolize herbicide molecules without damages to the plant, which is determined by the herbicide characteristics, rate used, and formulation. Therefore, the objective of this work was to evaluate the selectivity and efficiency of sulfentrazone-based formulations and herbicides in two tobacco production systems (conventional tillage system [CTS] and no-tillage system [NTS]).

Material and methods

Experimental area

The experiments were conducted in a commercial tobacco plantation area in the rural community of Garrafao, Imbuia, Santa Catarina, Brazil (27° 29’ 25” S, 49° 23’ 49” W, and 780 m a. s. l.) from September 2020 to March 2021. The soil of the area was classified as Haplustepts (Cambissolo Háplico Tb Distrófico) (Embrapa, 2018). The chemical analysis of the 0.0 to 0.2 m soil layer showed pH in water of 5.3, 47 % clay, 0.9 % organic material, P of 13.1 mg∙dm^-3, K of 213.8 mg∙dm^-3, Al of 0.5 cmolc∙dm^-3, cation exchange capacity of 19.03 cmolc∙dm^-3, and base saturation of 72.70 %. The climate of the region is Cfa, humid subtropical, according to the Köppen-Geiger climate classification, with a mean annual temperature of 19.1 °C and mean annual rainfall depth of 1,530 mm (Embrapa, 2012).

Experimental design and treatments

Two experiments were conducted under two different tillage systems (CTS and NTS), using a randomized block design with seven treatments and four replications. The treatments consisted of the following formulations: T1 = Boral^® 500 SC (400 g∙ha^-1 of sulfentrazone), T2 = PonteiroBR^® (400 g∙ha^-1 of sulfentrazone), T3 = Stone® (350 + 700 g∙ha^-1 of sulfentrazone and diuron, respectively), T4 = Boral^® 500 SC + Gamit^® 360 CS (792 g∙ha^-1 of clomazone), T5 = PonteiroBR^® + Gamit^® 360 CS, T6 = Stone^® + Gamit^® 360 CS, and T7 = control (manual weeding). In addition, a side strip adjacent to the experimental area was maintained without application of herbicides to assist in weed control evaluations.

The plots consisted of three tobacco planting rows with 5-m length, totaling an area of 6.5 m² per plot. The plot evaluation area consisted of disregarding 0.5 m from the ends, totaling 5.2 m². The total area of each experiment was 182 m².

The treatments (herbicides) were applied at pre-planting of tobacco and pre-emergence of weeds, on September 18, 2020, in an apply and plant system. The herbicides were applied using a CO₂ pressurized backpack sprayer equipped with four AD110.02 nozzles, at a constant pressure of 207 kPa, travel speed of 1.0 m∙s^-1, and an application rate of 200 L∙ha^-1. Weather conditions during the applications were monitored by a digital thermo-hygrometer-anemometer: air temperature of 23 °C, relative air humidity of 66 %, wind speed of 2.5 km∙h^-1, and wet soil.

Production systems and cultural practices

The area selected for experimenting CTS had been grown with black oats and ryegrass during the winter. It was subjected to CTS through two subsoiling operations 17 days before tobacco planting, and ridges were raised for planting tobacco seedlings. Regarding the area selected for conducting the experiment under NTS, soil tillage and ridging operations were performed as for the area under CTS, but in May 2020. Subsequently, black-oat seeds were broadcasted on the ground to provide soil cover during the winter and form straw for the following NTS. The black oat plants were desiccated with glyphosate (1,080 g∙ha^-1) on August 19, 2020, 30 days before planting tobacco seedlings, to form straw.

Tobacco seedlings (cultivar 416 of Souza Cruz) were transplanted manually to the soil on September 18, 2020, with spacing of 1.30 m between rows and 0.50 m between seedlings for a plant population of approximately 15,385 plants∙ha^-1.

Application of soil fertilizers at planting consisted of 760 kg∙ha^-1 of NPK 10-16-10 formulation. Topdressing application was performed 21 days after transplanting, using 470 kg∙ha^-1 of urea; and a second topdressing application was performed on November 02, 2020, using 1,340 kg∙ha^-1 of a formulation consisting of Chile saltpeter (25-16-84), NPK (10-16-10), and potassium chloride.

Removal of tobacco flower buds (topping) was performed on December 27, 2020. Only one application of insecticide was carried out to control aphids during the crop cycle, on January 15, 2021, using the insecticide Talstar^® 100 EC at rate of 25 mL∙ha^-1.

Evaluations

Phytotoxicity symptoms in tobacco plants and efficiency in weed control were evaluated visually at 14, 28, 42, and 56 days after application (daa) of the herbicide treatments. Phytotoxicity was assessed using a scale of grades from 0 to 100 %, where 0 represents absence of symptoms and 100 denotes the death of the plant (Kuva et al., 2016). The efficiency of treatments with herbicides in controlling weeds was evaluated 56 by counting randomly emerged weeds in an area of 1 m² per plot. Weed control and weed density were evaluated considering the entire infesting community, which consisted of Sida rhombifolia, Coronopus didymus, Ageratum conyzoides, Ipomoea triloba, Sonchus oleraceus, Raphanus raphanistrum, and Setaria parviflora.

The tobacco harvest was carried out in three phases: leaves from the lower third of the plants were harvested on December 12, 2020; leaves from the middle third of the plants were harvested on January 8, 2021; and leaves from the upper third of the plants were harvested on February 11, 2021. Tobacco yield per third of the plant (lower, middle, and upper thirds) was determined after drying the leaves. The total yield was obtained by summing the yields found for each third of the plant.

The tobacco leaves were dried in a masonry drying barn (LL model) with a total area of 49 m², consisting of two tiers, mechanically forced air circulation, and a brick furnace heated with wood. The temperature varied according to the stages of the drying process as follows: the temperature was maintained at 37.78 °C for 30 to 40 h for yellowing of the leaves; it was gradually increased from 37.78 to up to 54.44 °C over a period of 55 to 68 h for drying the leaves; and it was gradually increased from 54.44 to up to 65.56 °C over a period of 60 h for drying the stems. The dried tobacco was stored in a warehouse.

Statistical analysis

Considering the common structure of the experiments, the data were subjected to joint analysis. Thus, common factors were also considered, such as cultural practices, period in which the experiment was conducted, and edaphoclimatic conditions, among others. The order of magnitude of the residual mean squares from the individual analyses was followed (Banzatto & Kronka, 2008). The interaction between treatments and production systems was significant for most of the evaluated variables. Therefore, the decomposition of the interaction was performed by comparing means through Tukey's test. All analyses were based on a 5 % significance level (P < 0.05).

Results and discussion

Phytotoxicity symptoms were found only at 14 daa in plants under NTS, mainly for treatments with Stone^® (T3) and Stone^® + Gamit^® (T6); however, the injuries were considered very mild (2.3 and 2.5 % phytotoxicity, respectively) (Table 1). According to Oliveira (2011), initial chlorosis is a symptom commonly found in plants after application of diuron. Zacharias et al. (2021) found a decrease in the initial growth of soybean plants following the application of sulfentrazone + diuron; however, this effect was transient and did not affect the crop yield.

Table 1. Phytotoxicity (Phyto; %) in tobacco plants, efficiency in weed control (%), and weed density (Density; plants∙m^-2) under different tillage systems (CTS and NTS) and application of different sulfentrazone-based formulations and herbicide mixtures. Santa Catarina, Brazil (2020/2021 harvest season).

The application of formulations containing only sulfentrazone or mixtures containing sulfentrazone and clomazone did not cause phytotoxicity in tobacco plants (Table 1). Ritter et al. (2005) found mild injuries in tobacco plants varying from 0 to 3 % following application of sulfentrazone + clomazone, but with no effects on plant quality and yield. Clapp et al. (2022) found decreased growth for tobacco plants after applying sulfentrazone + clomazone; however, the final yield was higher than that found in the control (with no herbicide application).

According to Fisher et al. (2006), the tolerance of tobacco plants to sulfentrazone is due to the metabolization process. They found that tobacco plants, when recently transplanted, can rapidly metabolize this herbicide, as 66 % of the total absorbed by the leaves was metabolized within 3 h after application. However, the metabolization rate in leaves is higher when sulfentrazone is combined with clomazone.

Regarding weed control, there was a significant difference between the treatments and systems evaluated at 42 and 56 daa (Table 2). However, weed control higher than 96 % was found 42 and 56 daa for all herbicide treatments, and it was higher in plots under NTS (Table 1).

Table 2. Variance analysis for phytotoxicity (Phyto) in tobacco plants, efficiency in weed control, weed density (Density), yield of thirds of the plant (lower, middle, and upper), and total yield (entire plant) of tobacco plants grown under different tillage systems (CTS and NTS). Santa Catarina, Brazil (2020/2021 harvest season).

Bailey (2013) reported that applying combinations of two herbicides to control weeds in tobacco plantation areas provided more efficient control due to a broader spectrum of controlled weeds. He found that the tank-mixing application of sulfentrazone + clomazone was more efficient in controlling Ambrosia artemisiifolia, Cyperus esculentus, Digitaria sanguinalis, and Ipomoea hederacea in tobacco plantation areas. Haramoto and Pearce (2019) found lower weed density in areas where tobacco plants were grown on cover crop residues; however, the application of residual herbicides was needed to control these weeds. According to Haramoto et al. (2020), cover crop residues contribute to the reduction of light penetration into the soil surface layers, which prevents the emergence of weeds.

Regarding weed density, both systems (SCL and SLC) presented effective control, where the combination of PonteiroBR® + Gamit® (T5) showed the lowest value at 56 daa and was statistically different from Boral® 500 SC. (T1) (Table 1).

In general, weeds are more competitive than tobacco plants. In this sense, tobacco yields tend to decrease as weed density increases. Thus, an effective weed control system is necessary until the establishment of tobacco plants (Darwish et al., 2015). According to Bailey (2013), a high weed density in tobacco plantations can delay plant growth due to competition for light, water, and nutrients, negatively affecting the final yield and quality of tobacco plants. Clapp et al. (2022) reported that the application of herbicides for weed control in tobacco plantations reduces the need for manual labor and helps to decrease the supply of weed seeds to the soil seed bank.

Pre-emergent herbicides present residual activity in the soil, decreasing weed infestation over the crop cycle (Haramoto et al., 2020). Thus, the results of the present study showed that the herbicides used kept their residual activity in the soils under both tillage systems used (CTS and NTS). However, the effectiveness of residual herbicides is highly dependent on adequate soil moisture for activation; however, soil moisture is a factor that cannot be controlled in non-irrigated systems, as it is depends on climate conditions (Ritter et al., 2005).

Antoneli et al. (2016) evaluated soil surface moisture in tobacco plantation areas and found higher moisture contents in soils under NTS when compared to those under CTS. This may be attributed to the soil cover provided by straw, which contributes to the maintenance of high moisture levels, promoting an even distribution of moisture throughout the soil surface. Therefore, the higher weed control found in the present study for NTS (Table 1), compared to CTS, may be due to the greater soil moisture retention under NTS, resulting in better herbicide distribution and activity.

Regarding tobacco yield, NTS showed a significant difference from CTS, with higher yields in the upper third of the plants, resulting in higher total yield. However, no significant difference was found among the herbicide treatments (Table 3). This is a similar result to that found by Antoneli et al. (2016), who reported higher yield for tobacco plants under NTS than those under CTS. The herbicide treatments did not affect the tobacco plant yields, as they caused minimal phytotoxicity to tobacco plants; in addition, they provided excellent weed control, preventing competition between weed and tobacco plants for environmental resources. Thus, they presented similar yield to the control (manual weeding).

Table 3. Yield (kg∙ha^-1) of the lower, middle, and upper thirds of the plant and total yield (entire plant) of tobacco plants grown under different tillage systems (CTS and NTS) and subjected to application of different sulfentrazone-based formulations and mixtures. Santa Catarina, Brazil (2020/2021 harvest season).

Bailey (2013) found that herbicide applications resulted in higher tobacco yields compared to the control with no herbicide application and attributed this result to the mitigation of interference. The author also found no significant difference among the herbicides evaluated (sulfentrazone, clomazone, sulfentrazone + clomazone, pendimethalin, pendimethalin followed by sulfentrazone, pebulate, napropamide, and pebulate + napropamide); the tobacco yields found varied from 2,765 to 3,051 kg∙ha^-1 in the treatments with herbicides, whereas the control presented 2,406 kg∙ha^-1.

Tobacco yield in each plant profile is connected to the quality of the leaves grown. The leaves in the upper third of the plant provided higher yield under NTS and exhibited better quality due to a thick, sturdy laminar structure. However, the leaves in the lower third of the plant (bottom leaves) presented higher yield under CTS; they are the first leaves from the bottom up and have a thin laminar structure, followed by the leaves in the middle third with a medium sturdy laminar structure (Bendlin et al., 2020). This quality is evaluated in the marketing of tobacco leaves by classes and sale by weight (Ministério da Agricultura, Pecuária e Abastecimento, 2023).

Antoneli et al. (2016) found tobacco leaves that were 16.1 % longer under NTS compared to those under CTS; they attributed this result to soil moisture levels provided by NTS, which were higher in all evaluated months and varied less than those in CTS. Reichert et al. (2019a) found decreases in water, soluble and total P, and K losses when they combined planting on ridges under NTS; these losses under NTS were approximately four-fold lower than those found in soils without vegetation cover. Reichert et al. (2019b) found higher leaf area index, dry weight, and yield for tobacco plants under NTS compared to minimum tillage system. Therefore, soil management systems that involve soil cover plants and reduced tillage result in increased tobacco yield and improved quality of tobacco leaves.

Conclusions

This study is important for the tobacco sector as it provides an effective weed control method for use in tobacco cultivation, based on the use of different herbicides, thereby reducing soil disturbance and the demand for labor.

The herbicide Stone^® (350 + 700 g∙ha^-1 of sulfentrazone + diuron, respectively) applied alone or mixed with Gamit^® (792 g∙ha^-1 of clomazone) caused mild phytotoxicity in tobacco plants, characterized by transient symptoms that did not result in yield losses.

The evaluated sulfentrazone formulations applied alone or mixed with Gamit^® provided efficient weed control without significant damages to tobacco plants. The use of NTS provided higher efficiency in weed control and higher tobacco yields, thus preventing competition for environmental resources. This is mainly for use in the direct planting system, which provided greater efficiency in weed control with greater productivity and quality of tobacco leaves.