Grain yield and termite damage in rice by compaction on the furrow and seed treatment

The objective of this work was to determine the effects of soil compaction pressure on the sowing furrow and of seed treatments with insecticides on grain yield and on termite damage in upland rice plants under no-tillage, in the Center West region of Brazil. Three experiments were carried out, consisting of five seed treatments combined with three (first experiment) or four (second and third experiments) compaction pressures in a randomized complete block design, in a factorial arrangement. The seed treatments were: T0, without treatment; T1, imidacloprid+thiodicarb; T2, thiamethoxam; T3, carbofuran; and T4, fipronil+p yraclostrobin+thiophanate-methyl. The compaction pressures applied were: 25, 42, and 268 kPa, in the first experiment; and these same pressures plus 126 kPa in the second and third experiments. Seed treatment was effective in increasing rice grain yield, from 1,753 to 4,141 kg ha-1, especially in the lowest compaction pressure of 25 kPa. Soil compaction around 180 kPa in the sowing furrow results in higher rice grain yield and in fewer stems killed by termites. The use of soil compaction on the sowing furrow with seed treatment significantly increases rice grain yield under a no-tillage system.


Introduction
Rice is a grain included in the diet of half of the world's population, and this cereal is mostly grown under irrigation (Prasad, 2011).Growing rice under a no-tillage system (NTS) is an alternative to produce this crop in aerobic conditions (Nascente et al., 2013).According to Nokkoul & Wichitparp (2014), an area of approximately 20 million hectares is cultivated with upland rice, mainly in Latin America and Africa.The NTS, due to the benefits provided to the environment, such as greater conservation of water and increased soil fertility, organic matter content, and protection of the soil against erosion, had a significant global growth.The five countries with the largest areas under this system are the United States, Brazil, Argentina, Canada, and Australia, with 35.6, 31.8, 29.2, 18.3, and 17.7 million hectares, respectively (FAO, 2015).
However, under a NTS, upland rice crops have not had a good development (Nascente et al., 2011).In this system, the 0.00-0.05-msoil layer usually shows high porosity, which is attributed to the following factors: action of the seeder discs; highest concentration of roots and organic matter; greater biological activity; and more cycles of soil wetting and drying (Collares et al., 2006;Reichert et al., 2009), which can hinder the contact of the rice seed with the soil, reducing plant emergence.
In this context, the compaction on the sowing furrow performed by the seeder press wheel improves ground contact with the seeds, allowing them to absorb water earlier and to germinate faster (Modolo et al., 2007).This practice increases upland rice grain yield under no-tillage (Pinheiro et al., 2016a(Pinheiro et al., , 2016b)).
Termites, such as Procornitermes araujoi, P. triacifer, and Syntermes molestus, occur in most upland rice fields established on soils in the Cerrado (Brazilian savanna), causing significant losses in crop grain yield, and are also the main reason insecticides are used to treat seeds before sowing (Ferreira et al., 2007).Under a NTS, the problem is compounded because of the straw accumulation on soil surface, increasing food availability for termites, which can cause total losses in rice production (Barbosa et al., 2009).In addition, in this system, there is no reduction in the population of these insects by exposure, crushing, and destruction of galleries, as done in conventional tillage (plowing once and harrowing twice).
Chemical control of termites has been the most widely used and the most effective control method (Verma et al., 2009).However, the detrimental effect of insecticides on the environment cannot be ignored.An alternative to the use of insecticides is soil compaction on the sowing furrow.Ferreira et al. (2007) reported that the termite attack on upland rice grown under notillage was lower with soil compaction on the sowing furrow, in comparison with no compaction, and that the benefit of this practice was reinforced when combined with seed treatment with insecticides.
Despite the importance of the study of soil compaction on the sowing furrow for rice development and the reduction of termite attack, there is still little scientific evidence on this subject.
The objective of this work was to determine the effects of soil compaction pressure on the sowing furrow and of seed treatments with insecticides on grain yield and on termite damage in upland rice plants under no-tillage, in the Center West region of Brazil.

Materials and Methods
The experimental area was located in the municipality of Santo Antônio de Goiás, in the state of Goiás, in the Center West region of Brazil (16º27'28"S, 49º19'52"W, at 823 m above sea level).The climate is a tropical savanna, Aw according to Köppen's classification.There are two distinct seasons, usually dry from May to September and rainy from October to April, with average annual rainfall of 1,485 mm.During the experiment, temperature and rainfall data were recorded (Figure 1).
Three experiments were carried out with upland rice crop under no-tillage.In the first one, rice was sown on January 27, 2011, on desiccated straw of Urochloa brizantha, after nine years of pasture.The second experiment was sown in another area, on December 8, 2011, on desiccated rice straw from the previous year, whereas the third experiment was sown on January 24, 2013, on the desiccated straw of the second experiment.
The experiments were conducted in a randomized complete block design, in a factorial arrangement, with three replicates, and consisted of five seed treatments combined with three (first experiment) or four (second and third experiments) soil compaction pressures on the sowing furrow (CP).Based on the results of the first experiment, it was decided to increase the soil compaction pressures to obtain more information.The seed treatments were: T1, imidacloprid+thiodicarb, 52.5 + 157.5 mL a.i.per 100 kg seed; T2, thiamethoxam, 140 g a.i.per 100 kg seed; T3, carbofuran, 527 mL a.i.per 100 kg seed; and T4, fipronil+pyraclostrobin +thiophanate-methyl, 62.5 + 0.625 + 50.625 mL a.i.per 100 kg seed.All treatments were compared with a control (T0) without seed treatment.The soil compaction pressures applied in the first experiment were: 25 kPa, provided by the original "V" press wheel of the seed drill; 42 kPa, provided by the mass of 96.6 kg of a wheel adapted in the seed drill; and 268 kPa, provided by the mass of 635 kg on the front axle of a MF 5275 4x2 tractor (Massey Ferguson, AGCO, Duluth, GA, EUA) with 7.50-16 F2 directional vehicular tires with three treads (smaller contact area and high pressure on the ground).These same CP were applied in the second and third experiments plus the pressure of 126 kPa, provided by the mass of 635 kg on the front axle of the same tractor with 7.50-16 bald directional vehicular tires (no slots, larger contact area, and reduced pressure on the ground).The compaction pressures were estimated based on the weight of the wheels or tractor and on the contact area of the wheels or tires with the ground.
The second experiment was conducted in 10x2.25-mplots, with the same cultivar.Sowing was done with the same seed drill, and sowing fertilization was also performed with 17 kg ha -1 N as urea, 102 kg ha -1 P 2 O 5 as triple superphosphate, and 51 kg ha -1 K 2 O as potassium chloride.Nitrogen topdressing fertilizations as urea (42 kg ha -1 ) were performed at 18 and 35 days after rice emergence.
The third experiment was carried out on the same plots as the second one, with the same cultivar, seed drill, and sowing fertilization.Nitrogen topdressing fertilizations as urea were done at 1 (20 kg ha -1 ), 20 (40 kg ha -1 ), and 28 (28 kg ha -1 ) days after rice emergence.
In all trials, the same tillage practices were applied to keep the area free from weeds, diseases, and Moreover, the usable area of the plot was composed of three central rows of rice, disregarding 0.50 m on each side.
In the second and third experiments, the soil resistance to vertical penetration (PR) was determined through a penetrograph, penetroLOG-PLG 1020 model, (Falker, Porto Alegre, RS, Brazil) at the 0.00-0.20-mdepth, at three points in each plot, one in each useful line.The data were obtained in kPa, and average values of PR were calculated for the 0.05-m layers until the 0.20-m depth.Sampling was performed one day after rain, with soil moisture close to field capacity.The data was presented as the cone index to the depth of 0.20 m, considering the mean value of cone penetration resistance.
Rice harvesting was done after physiological maturity, with cutting by hand in the usable area in each plot.Plots were assessed regarding grain yield, which was determined by weighing the harvested grain of each plot, corrected to 13% of water content and converted to kg ha -1 .In all experiments, the number of stems killed by termites in 1 m of row (SK), in three predetermined points on the centerlines of the plots, was evaluated throughout the crop season.
Data were subjected to analysis of variance using the Sisvar statistical software (Universidade Federal de Lavras, Lavras, MG, Brazil), and means were compared by Tukey's test, at 5% probability.Regression analysis was also performed to determine the effects of CP and SK on grain yield.

Results and Discussion
A significant effect was observed on seed treatment x compaction pressures, in terms of rice grain yield in the first experiment (Table 1).In the other experiments, there was a significant single effect of seed treatment and of compaction pressures on grain yield.Regarding stems killed by termites, the effect of seed treatment was significant only in the first experiment, whereas the effect of compaction pressures was significant in the second and third ones.
The increase in grain yield with seed treatment with insecticides, when compared with the untreated ones, is probably due to the reduction in crop damage by soil insects.Ferreira et al. (2007) found that seed treatment with insecticides reduced plant mortality caused by termites and the lesser cornstalk borer (Elasmopalpus lignosellus), increasing grain yield.Pinheiro et al. (2016b) also reported that seed treatment provided greater control of termites in upland rice crop than no treatment.However, Ferreira et al. (2007) pointed out that the insecticide carbofuran, although effective in the control of sucking insects that attack rice in the early stages of plant development, had low efficiency in controlling termites.This is an indicative that differences regarding grain yield among seed treatments with insecticides are probably due to their effect on termites.
In the first experiment, at the lowest compaction pressure of 25 kPa, the T0 treatment had the lowest grain yield and differed from all the other ones (Table 2); however, there was no significant difference among the other seed treatments.At the compaction pressure of 42 kPa, the T1 and T4 treatments provided the highest grain yield, differing from T0 but not from the other seed treatments.At the highest compaction pressure of 268 kPa, there was no effect of seed treatment on grain yield.It should be noted that the absence of seed treatment favored higher termite damage, resulting in a greater number of stems killed, which differed from treatments T1, T3, and T4 (Table 3) and affected negatively grain yield (Figure 2).In T1, T3, and T4, there was no significant effect of compaction pressures on grain yield; however, under the T0 and T2 treatments, grain yield increased linearly with compaction pressures (Table 4).
In the second experiment, the lowest grain yield was obtained in T0 (2,158 kg ha -1 ), which differed from the other treatments (Table 3).The only exception was T3, which showed similar results to T0.In the third experiment, grain yield in the T0 treatment was significantly lower than that obtained in T4.Although the number of stems killed was greater in absolute values in the second and third experiments under T0, there was no significant difference among seed treatments regarding this variable (Table 3).The increase in the number of killed stems could be explained by the fact that the second and third experiments were performed after the rice crop, which is a known preferable food for termites (Ferreira et al., 2007).
The increase in compaction pressures contributes to control termites (Pinheiro et al., 2016a) and is possibly why, in general, there were no differences among seed treatments at higher compaction pressures.According to Radford & Allsopp (1987), soil compaction over the rice rows may: squash large insect pests near the seed, killing or injuring them; restrict the movement of insects through the soil; improve soil water supply, enabling seedlings to escape pest damage by faster germination and growth; or reduce carbon dioxide diffusion from the seed/seedling, by reducing soil porosity -it should be highlighted that carbon dioxide attracts several species of wireworm.
In the second and third experiments, grain yield showed a quadratic response increase to compaction pressures (Table 5), with the maximum values being reached with the soil pressures of 181.6 and 177.9 kPa, respectively.The number of stems killed also showed a quadratic response to compaction pressures, but with minimum values of 195.7 and 186.2 kPa, respectively.
The effect of compaction pressures on grain yield depends on the intensity of the load applied.Modolo et al. (2007) found that low loads on the sowing furrow decreased the contact between the soil and the seed by limiting the amount of water that reaches seed surface, reducing plant emergence; however, high loads can   T0, without seed treatment; T1, imidacloprid+thiodicarb, 52.5 + 157.5 mL a.i.per 100 kg seed; T2, thiamethoxam, 140 g a.i.per 100 kg seed; T3, carbofuran, 527 mL a.i.per 100 kg seed; and T4, fipronil+pyraclostrobin+ thiophanate-methyl, 62.5 + 0.625 + 50.625 mL a.i.per 100 kg seed.cause surface crusting, reducing the penetration of oxygen, so that the plant needs to consume more energy to emerge.Pinheiro et al. (2016a) also evaluated upland rice under no-tillage and reported a favorable effect of compaction pressures on grain yield.
Since the third experiment was conducted on the straw of the second one, it can be concluded that the continuous cultivation of rice in the same area increased damage by termites and reduced grain yield (Table 3 and  Figure 2).This lower grain yield in the third experiment, in comparison with the second one, is in alignment with the results obtained by several other authors, who do not recommend the continuous cultivation of upland rice in the same area due to the sharp fall in productivity (Fageria & Baligar, 2003;Nie et al., 2009;Godoy et al., 2013).This decline in grain yield has been attributed to allelopathic effects (Fageria & Baligar, 2003), which are any direct or indirect harmful effects of a plant on another one through the production of chemical compounds released into the environment.
The greater accumulation of exudates released by rice plants and the larger number of glyphosate applications in the area of the second rice crop may have altered the activity and the composition of the microbial community, contributing to the lower rice grain yield, as observed by Godoy et al. (2013).Furthermore, the continuous cultivation of rice increased termite infestation in the area, also increasing the number of dead stems and reflecting in grain yield drop, as found in the present work.
Penetration resistance increased gradually from the surface to the 0.20-m depth with increasing compaction pressures (Figure 3).In the second experiment, the penetration resistance value reached around 3,200 kPa at the two higher compaction pressures studied.For these same pressures and for 42 kPa, the value of penetration resistance was around 1,850 kPa in the third experiment.
The increase in penetration resistance with the increase in depth is probably due to the lack of soil disturbance and the presence of machinery traffic, as also reported by Cunha et al. (2011).However, the increase in penetration resistance with increasing compaction pressures must not have damaged the rice crop, since the penetration resistance values that limit root growth and crop yield were close to 3,000 kPa (Reinert et al., 2008).Moreover, according to Tormena et al. (2007), this value can reach 3,500 kPa in soil cultivated under long-term no-tillage, justified by the presence of continuous and effective biopores.Therefore, according to the obtained results, it can be inferred that the use of soil compaction on the sowing furrow can contribute to increase rice grain yield without damaging rice root development.

Conclusions
1. Seed treatment is effective in increasing upland rice grain yield, from 1,753 to 4,141 kg ha -1 , especially in the lowest compaction pressure of 25 kPa.
2. Soil compaction around 180 kPa on the sowing furrow results in higher rice grain yield and fewer stems killed by termites.

Figure 1 .
Figure 1.Maximum, minimum, and average temperatures (T), and rainfall recorded in the experimental field during the experimental period.

Figure 2 .
Figure 2. Upland rice grain yield as a function of number of stems killed by termites in 1 m of row. 1 st Exp, 2 nd Exp, and 3 rd Exp, first, second, and third experiments carried out.T0, without seed treatment.

Figure 3 .
Figure 3. Penetration resistance as a function of compaction pressure on the sowing furrow, in the second (A) and third experiments (B).

Table 1 .
P-values of the F tests for source of variation of upland rice grain yield (GY) and number of stems killed by termites in 1 m of row (SK) in the three experiments carried out.

Table 3 .
Upland rice grain yield and number of stems killed by termites in 1 m of row affected by seed treatment(1).

Table 4 .
Regression equation for grain yield (GY) as a function of compaction pressure on the sowing furrow (CP), in the first experiment.

Table 5 .
Regression equations for grain yield (GY) and number of stems killed by termites in 1 m of row (SK) as a function of compaction pressure on the sowing furrow (CP), in the second and third experiments (n = 60).