Rice (Oryza sativa L.) has great economic, social, and food importance due to its daily consumption habit. The annual consumption per capita of rice in Brazil varies between 25 and 50 kg inhabitant⁻¹ year⁻¹ (FAOSTAT, 2011). Rice grain is an excellent source of carbohydrates, proteins, minerals (mainly phosphorus, iron and calcium) and B vitamins such as B₁ (thiamine), B₂ (riboflavin) and B₃ (niacin). Rice was cultivated on a total area of 1.7 million hectares with production of 10.6 million tons and an average yield of 6.300 kg ha⁻¹ in 2019–2020 in Brazil (Conab, 2020). Therefore, it is important to adopt sustainable practices and proper management to ensure high quality and production of rice in tropical savannah.

Soil compaction in no-tillage system (NTS) mostly occurs in superficial soil layers due to heavy traffic of machinery and implements in the soil with high moisture content or by mobilizing soil only in sowing furrow (Carof et al., 2007; Francaviglia et al., 2015). The heavy intervention of machinery in the absence of adequate agricultural planning for cultivation of cover or successive crops contributed to the intensification of compacted surface soil layers in NTS that could disrupt soil structure with affecting soil fertility and crop productivity (Franchini et al., 2012). Soil compaction increased density and root penetration resistance, reduced total porosity, macro-porosity, and altering water and nutrients dynamics (Tokura et al., 2017). Therefore, some sustainable and environmental friendly soil management practices like mechanical scarification with subsoiler and scarifiers while biological scarification with cover crops cultivation are needed to adopt to counter the harsh effects of soil compaction in long term NTS.

Mechanical intervention is carried out by means of scarifiers or subsoilers with cutting discs in front of stems that resist crop residues to incorporate into the soil. However, long-term effects of mechanical soil scarification is altering with time limit, ranging from few months (Seidel et al., 2018; Nazari et al., 2021) to few years (Boreta Junior et al., 2021), relaying on the redisposition of soil particles as a result of soil type, weather conditions, machines and implements intervention, and predominant management practices in particular production system (Silva et al., 2017). Soil compaction has a direct on affect physical and mechanical characterizations of soil which consequently impair plant growth and development. In general, soil compaction impairs water and nutrients uptake by limiting root length and penetration that all ultimately lead to poor plant growth and yield (D’Or and Destain, 2016). Scarification breakdown compacted soil layers and generating macro-pores to improve soil permeability and facilitate root penetration and corrective movement.

Thus, management of plant crops in the no-tillage system, such as cultivation of cover crops or in intercropping systems is a promising alternative to increase plant biomass productivity and nutrient accumulation in no-tillage system (Pereira et al., 2016; Chieza et al., 2017). The plants of Poaceae family such as Pennisetum glaucum, Urochloa ruziziensis, U. brizantha and Avena sativa are fast establishing species and capable of high biomass production and nutrient cycling (Teixeira et al., 2012; Mauro et al., 2015; Pereira et al., 2016). The introduction of Fabaceae family species in production systems, alone or intercropped with Poaceae family species have the ability to increase soil nitrogen (N) availability from biological nitrogen fixation (BNF) and mineralization (Stagnari et al., 2017). In addition, cover crops are important for regulating soil sustainability, nutrients recycling and providing several benefits to the crops of economic interest.

The inclusion of cover plant species (leguminous and/or grasses) have the capability to alleviate soil compaction by acting as barrier between machinery and soil as well as increasing mineralization and nutrient recycling. In addition, natural BNF capacity of leguminous plants allow the reduction of N supply by up to 23% in the first cropping season (Adeux et al., 2021), while increasing grain yield in the absence of N fertilization in coverage (Mahama et al., 2020). The use of Poaceae species as cover crops provide longer period to cover soil with their residues due to thier high C:N ratio and organic matter mineralization (Raza et al., 2019). The combination of mechanical scarification with cover crops reduce compaction of topsoil layer, increasing root development and penetration, yield and quality of successive crops (Melkonian et al., 2017; Sarto et al., 2018; Fin et al., 2018; Fin et al. al., 2018; Adeux et al., 2021).

Considering the above discussion of maintenance and alterations in management practices in NTS and a diverse role of cover crop for successive crops, the hypothesis adopted that would mechanical soil scarification and cover crops in NTS have positive influence on grain yield and quality of rice. The objective of the study aimed to evaluate the effects of mechanical soil scarification associated with previous and successive cultivation of cover crops on grain yield and industrial quality of upland rice in tropical Cerrado of Brazil, being no-tillage system for 12 years.

Rice seedlings in both cropping seasons were emerged uniformly on the sixth day after sowing (DAS), reached to full flowering at 82 days after emergence (DAE) and harvested at 106 DAE. There was no problem of plant lodging in both cropping seasons. The cultivation of predecessor cover crops positively influenced plant height in 2013/2014 cropping season (Table 1). The cultivation of U. ruziziensis and C. cajan as cover crops increased plant height by 4.7 and 2.8% in relation to C. juncea, which was noted with dwarf plants.

The interactions of cover crops and mechanical soil scarification were significant for number of panicles m⁻² in both cropping seasons while the interactions for total number grains panicle⁻¹, number of filled grains panicle⁻¹ and empty grains panicle⁻¹ were significant only in 2013–2014 cropping season in no-tillage system (Table 1).

Number of panicles m⁻² of rice were increased with cover crops cultivation associated or not with mechanical soil scarification in no-tillage system (NTS) in both cropping seasons (Table 1; Figures 2A,B). Highest number of panicles m⁻² in 2012/2013 were observed in fallow treatments which was statistically similar within the treatments of C. cajan in association with mechanical soil scarification as compared to other treatments (Figure 2A). In addition, number of panicles m⁻² in second cropping season were increased with mechanical soil scarification in the treatments incorporated with crop residues of C. cajan which were statistically at per with P. glaucum and fallow area. The lowest number of panicles m⁻² were observed with residues of C. juncea in mechanical soil scarification in NTS (Figure 2B). However, treatments without scarification were observed with greater number of panicles m⁻² of rice when previously cultivated with P. glaucum and C. juncea and lowest with U. ruziziensis in first cropping season (Figure 2A). In second cropping season, previous cultivation of P. glaucum produced higher number of panicles m⁻² which was statistically similar with C. juncea and U. ruziziensis in non-scarifies treatments while the lowest was noted within fallow treatments (Figure 2B).

The mechnical scarification in fallow crop treatments were oberved with greater total number of grains panicle⁻¹ which was statistically similar with U. ruziziensis treatments while cultivation of P. glaucum, C. juncea and C. canjan as a cover crop without scarification produced greater total number of grains panicle⁻¹ in relation to fallow in second cropping season (Figure 2C). The lowest total number of grains panicle⁻¹ in rice were observed with cultivation of P. glaucum, C. juncea and C. canjan in scarified treatments while without scarification in fallow treatments (Figure 2C).

The number of filled grains panicle⁻¹ in 2013/2014 were higher with residues of U. ruziziensis and fallow treatments in mechanical soil scarification. The cultivation of P. glaucum and C. juncea as cover crops in non-scarified treatments were observed with higher number of filled grains panicle⁻¹ (Figure 2D). The lowest number of filled grains panicle⁻¹ were observed with C. juncea in scarification and in fallow without scarification in relation to other treatments (Figure 2D).

The interaction of mechanical soil scarification and cover crops for empty grains panicle⁻¹, 100-grain weight and hectoliter mass were significant only in 2013/2014 cropping season (Tables 2; Figure 3). The highest number of empty grains in mechanical soil scarification was observed with treatment of C. juncea as a cover crop which was statistically similar with fallow treatments. The lowest empty grains panicle⁻¹ were observed with the effect of scarification in soils previously cultivated with P. glaucum as compared to other treatments (Figure 3A). In addition, treatments of C. canjan in non-scarified soils produced higher number of empty grains panicle⁻¹ which was statistically not different from the treatments of P. glaucum and U. ruziziensis. The lowest empty grains panicle⁻¹ were observed with fallow treatments in non-scarified soil (Figure 3A).

The treatments with U. ruziziensis and P. glaucum in NTS with mechanical soil scarification had greater 100-grains weight. The lowest grains weight in mechanical soil scarification was observed with previous cultivation of C. canjan as cover crop in NTS (Figure 3B). In addition, 100-grains weight of rice was increased with C. cajan and C. juncea without mechanical soil scarification in NTS. However, without scarification, low grains weight was noted within fallow treatments (Figure 3B).

Hectoliter mass of rice grains was increased with mechanical soil scarification in previous cultivation of P. glaucum under NTS. (Figure 3C). The mechanical soil scarification within previous cultivated U. ruziziensis in NTS showed potential for increasing hectoliter mass of rice in 2013/2014 cropping season. The lowest hectoliter mass in mechanical soil scarification was observed with previous cultivation of C. canjan (Figure 3C). The non-scarification within treatments of P. glaucum as cover crop also increased hectoliter mass of rice while the lowest was noted with C. canjan as a cover crop in relation to other treatments (Figure 3C).

The management with mechanical scarification and cover crops significantly influenced grain yield of upland rice in 2012/13 cropping season while treatments effect in 2013/14 and their interactions in both seasons were not significant (Table 2). Mechanical soil scarification increased grain yield of upland rice by 12.2 and 2.9% as compared to without scarification in 2012/13 and 2013/14 respectively. In addition, previous cultivation of C. juncea and C. canjan have increased grain yield of upland rice by 12.3 and 8.5% as compared to fallow treatments in 2012/13 and 2013/14 cropping season.

Milling yield was significant only with mechanical soil scarification in 2012/13 cropping season (Table 3). The operation of mechanical soil scarification improved milling yield by 1.8% in relation to without mechanical soil scarification. The whole and broken grains yield of upland rice in both cropping seasons were not influenced by mechanical soil scarification and cover crops, and their interactions grains in the two harvests, there was a significant effect only for the milling yield in the 2012/2013 harvest (Table 3).

Based on the results, soil management with mechanical scarification and cover crops under sprinkler irrigation are promising alternatives to develop sustainable agriculture system in NTS of low-altitude Cerrado of Brazil.

The superficial mechanical scarification of soil provided greater environment for development of rice plants in all cover crops cultivation. This is due to the fact that mechanical scarification of soil allowed the plants to develop greater roots system (Chieza et al., 2017). Since, the current experiment was conducted in soils being under no-tillage system (NTS) for 12 years and accumulated soil particles of low porosity due to traffic heavy machines in crop cultivation, treatments application and harvesting. This high degree compaction of soil surface layer may harm soil profile by limiting penetration of tap and secondary roots in depth as well as reducing water and nutrients absorption, that all hinder growth and productivity of crops (Seki et al., 2015; Monteiro et al., 2017).

The highest number of panicles m⁻², number of total grains and number of filled grains panicle⁻¹ of upland rice in 2013/2014 cropping were observed with C. juncea and P. glaucum as cover crops without mechanical soil scarification and in fallow with mechanical soil scarification in NTS (Figure 2). Reis et al. (2017) also reported that inclusion of C. juncea as a cover crop in rice cultivation system had increased number of panicles and panicle mass in a no-tillage system without mechanical soil scarification. The incorporation of crop residues by rotary tiller in NTS has improved physical and chemical characterizations of soil, even though resulted in greater grain yield of rice due to greater organic matter retention (Tang et al., 2020). Similarly, Klein et al. (2008) indicated that number of grains spikelet⁻¹, spikelets spike⁻¹ and hectoliter mass were increased with mechanical soil scarification in a NTS than without mechanical soil scarification. Therefore, the present results also showed that mechanical soil scarification in the treatments with U. ruziziensis and P. glaucum as cover crops in NTS were observed with greater 100-grains mass and hectoliter mass of upland rice (Figures 3B,C). The might be due to the role of mechanical scarification in NTS that reduces soil compaction, sustaining organic matter, increasing infiltration and retention of water in deep soil layers and also increasing soil aeration (Monteiro et al., 2017).

The operation of mechanical soil scarification and cover crops influenced grain yield of upland rice in the first cropping season (Table 1). Soil management with mechanical scarification has increased grain yield of rice by 12.2% in relation to treatments without mechanical scarification. Milling yield was 1.8% higher with mechanical scarification in no-till soil as compared to non-scarified soil (Table 2). These results can be explained by the fact that mechanical soil scarification increases total porosity and macro-porosity, reduces soil density by disrupting compaction of soil surface layers (Monteiro et al., 2017). This might be due to the reduction of penetration resistance of roots to mechanical scarification of soil which improved root system (Valentine et al., 2012) with greater absorption of water and nutrients (Stagnari et al., 2017; Sarto et al., 2018). The greater efficiency of soil exploration by roots has a direct effect on water and nutrient transport from the roots to aerial part of the plants that has ultimately increasing photosynthesis and transport of photo-assimilates to grains Fin et al., 2018), leading to better grains filling. However, this effect decreases with time as traffic of heavy agricultural machinery in NTS lead to the compaction of soil due to natural reorganization of soil particles (Troccoli et al., 2015). Nunes et al. (2014) described that positive effects of mechanical scarification on physical attributes of soil and root development of corn didn’t persisted for more than 18 months. Therefore, the effect of frequently used machinery on soil can be mitigated by the intorduction of cover crops, a large amount of straw residues provided by grass species. In addition to the ability of some cover crop species to reduce surface soil compaction, they also allow to increase crop productivity (Sarto et al., 2018; Ogilvie et al., 2021).

The effects of cover crops in 2012/13 cropping season indicated that cultivation of C. cajan provided greater grain yield of upland rice than P. glaucum, U. ruziziensis and fallow treatments (Table 2). This might be due to greater decomposition potential of C. cajan that consequently release nutrients by mineralization for the crop in succession, in this case rice is successive crop. It was reported that P. glaucum straw can reduce nitrogen supply by biological N fixation and mineralization of P. glaucum as crop residue thus contribute to higher grain yield of rice (Reis et al., 2017). In other case it was indicated that U. ruziziensis was the best option of cover crop before rice cultivation however, low grain yield was due to the deficiency of water during pre-flowering stage (Pacheco et al., 2013), different from the experiment in question as it was irrigated with sprinkler irrigation system.

The integrated use of grasses as cover crops and mechanical scarification provided greater soil exploration by soybean roots which has consequently increased grain yield (Fin et al., 2018). The crop residues of C. juncea are very similar to poaceae with relatively high persistence due to high lignin content in stems that reduce soil temperature and water loss therefore, promoting root-shoot growth of crops in succession (Valentine et al., 2012; Nascimento et al., 2019). The use of U. ruziziensis as cover crop improved fertilizers use efficiency such as macronutrients and managing to provide high amounts of these nutrients for subsequent crops (Janegitz et al., 2016). In addition, Pacheco et al. (2011) reported that highest rice grain yield in no-tillage system was obtained after cultivation of P. glaucum and U. ruziziensis. The introduction of legumes as cover crops can reduce N fertilization by 23% as well as increasing transportation and accumulation into the shoot and grains of sorghum (Mahama et al., 2020; Adeux et al., 2021). Pedrinho et al. (2018) indicated that using grasses as cover crops increased grain yield and reduced N fertilization by 50 kg ha⁻¹ while legumes as cover crop has reduced N fertilization by 100 kg ha⁻¹. Thus, use of cover crops in agriculture system are able to reduce nitrogen fertilizers as well as increasing profitability and economy along sustainable production of rice.

Mechanical soil scarification and cover crops in no-tillage system has positive impact on growth, yielding components, yield and quality of upland rice in succession. The cultivation of C. cajan as a cover crop in mechanical soil scarification provided greater number of panicles m⁻² of upland rice.

The mechanical soil scarification in fallow provided greater number of total grains and filled grains of upland rice. The cultivation of C. cajan as covers crop increased number of empty grains in without scarification treatments. Mechanical scarification within previous cultivation of U. ruziziensis and P. glaucum provided higher 100-grains mass whereas greater hectoliter mass was noted withot scarification in cultivation of U. ruziziensis as a cover crop.

Mechanical soil scarification and Cajanus cajan as cover crop increase grain yield and milling yield of upland rice in tropical conditions of Brazil. The grains quality of upland rice was less effected by treatments as compared to productive attributes.