SOIL PROPERTIES AND CROP YIELD UNDER DIFFERENT TILLAGE METHODS FOR RAPESEED CULTIVATION IN PADDY FIELDS

A two-year research was conducted to investigate the effect of different tillage methods on some soil physical characteristics and crop yield in rapeseed cultivation after rice harvesting. Five tillage treatments including: (i) using rotavator, once to depth of 10–15 cm (T1), (ii) using rotavator, twice to depth of 10–15 cm (T2), (iii) using moldboard plow to depth of 25 cm + rotavator, once to depth of 10–15 cm (T3), (iv) no-till planting through removing rice stubbles from plots (T4), and (v) no-till planting without removing rice stubbles from plots (T5), were evaluated under randomized complete block design (RCBD) in three replications. The biannual results revealed that the effect of tillage methods was significant (p<0.01) on soil bulk density, surface residues after tillage, dry mass of weeds, seed germination, and grain yield. T2 and T3 made considerable reduction in soil bulk density compared to other treatments for the 15to 30-cm tillage depths. In T1, T2, T3, and T4, surface residues after tillage decreased in comparison with T5 by up to 35.37, 50.71, 69.92, and 75.75%, respectively. Having 71.48 g m -2 , T5 had the maximum dry mass of weeds while T3 had the minimum one with 37.50 g m -2 . Means comparison represented that in T2 and T3, seed germination reached the shortest length of 6.4 days in average. The highest and lowest grain yields were acquired in T3 (1,571 kg ha -1 ) and T5 (1,339 kg ha -1 ), respectively. Statistically, there was no significant difference between T1 (1,432 kg ha -1 ) and T2 (1,537 kg ha 1 ) compared with T3 in terms of grain yield.


Introduction
Land preparation is one of the fundamental steps in producing farm products.In paddy fields, because of soil physical and moisture content conditions as well as rice stem residues, preparing land for second crops encounters studious and laborious problems.Among them, sticking soil particles to implements, stranding plow bottom, and over-slippage of tractor drive wheels are remarkable.Therefore, there will not only be inappropriate plowing quality, but also it makes expenditures rise severely in such conditions.Hence, in order to provide a favorite bed for plant growth and reduce production costs, it is vital to develop suitable tillage methods in rapeseed cultivation as a second crop after rice.
There have been few studies concerning the effect of tillage methods on soil physical characteristics and yield of rapeseed which are discussed here.Bonari et al. (1995) compared conventional tillage (25-cm plowing depth) with minimum tillage (10-to 15-cm disc harrowing depth) for winter rapeseed production in a very sandy soil.They concluded that rapeseed grain and biomass yields did not differ significantly in both systems.Torabi et al. (2008) investigated the effect of tillage and date of sowing on rapeseed yield and soil physical and biological characteristics.Their results indicated that tillage and date of seed sowing had a significant effect on grain yield and soil physical properties.
An investigation conducted by Fooladivand et al. (2009) demonstrated that the highest grain yield was associated with conventional tillage method while notillage method had the lowest one because of high sensitivity of rapeseed to appropriate seed bed preparation and tiny seed size.Chiriac and Jitareanu (2011) carried out trials to examine the effects of tillage on soil properties and yield of rapeseed.They studied soil bulk density, total porosity, aggregate water stability, and mean weight diameter under tillage treatments.Their results indicated significant differences between tillage treatments in terms of soil physical parameters.Also, in time, they lost their significance because soil tended to revert to initial state before tillage due to relatively high rainfall during that period.Iriarte et al. (2011) examined the effects of different fertilization and tillage practices on environment and energy demand of rapeseed production.In their study, they used conventional tillage and zero tillage in four mineral fertilization experiments at a local level.They found out that two tillage systems had significant impacts on environmental profiles while the energy demand was not affected by tillage systems significantly.Also, it was revealed that other sources of fertilizers should be assessed to reduce environmental impacts.
The possibility of adding a rapeseed cropping season after late rice harvest was investigated by Wang et al. (2011) in order to extend direct seeding rapeseed under no-tillage pattern.In their experiments, the effects of plant density and nitrogen application rate were assessed at different growth stages.They found out that in the case of no-tillage and direct seeding pattern in the double rice cropping area, both nitrogen application rate and planting density should be considered simultaneously to obtain high quality of plant population and high yield of rapeseed.
The influence of different land preparation methods on yield of irrigated rapeseed after corn was evaluated by Taghinezhad et al. (2012).Their results showed that tillage method had a significant effect on soil bulk density for the 10-to 20-cm plowing depth.Chiriac et al. (2012) examined three tillage methods: moldboard plow (MT), chisel (CT), and disc harrow (DT) and 50 rapeseed cultivars on rapeseed yield to indicate the most favorite tillage treatment and highyielding cultivar.The results demonstrated that MT was the best treatment in terms of rapeseed yield.But, there was no significant difference (p<0.05) in terms of mean yield per cultivar.Rapeseed yield was only influenced significantly by cultivar in the first year of experiments.They found no significant difference for the entire study period.The effect of three tillage systems: moldboard plow (MT), chisel (CT), and disc harrow (DT) and 50 rapeseed cultivars was evaluated in terms of yield and economic efficiency (Chiriac et al., 2013).They expressed that MT, in spite of having the highest costs, was the most efficient treatment because of having the highest yield outcomes.
This study aimed to investigate the effect of different tillage methods on some soil physical characteristics and grain yield of rapeseed as a second crop after rice harvest.

Material and Methods
The experiments were conducted at research paddy fields of Rice Research Institute of Iran, Rasht, Iran, during farming seasons of 2009-2010 and 2010-2011.Rice harvesting was performed in the late August like past years and rice stubble residues roughly 45 cm high were left over across the field by the late September (beginning of tillage operations).The experiments were laid out in randomized complete block design (RCBD) with five treatments and three replications.The tested treatments consisted of: (i) using rotavator, once to depth of 10-15 cm (T 1 ), (ii) using rotavator, twice to depth of 10-15 cm (T 2 ), (iii) using moldboard plow to depth of 25 cm + rotavator, once to depth of 10-15 cm (T 3 ), (iv) no-till planting through removing rice stubbles from plots (T 4 ), and (v) no-till planting without removing rice stubbles from plots (T 5 ).In treatment T 4 , rice stem residues were cut off by labors and transferred to outside of the field.In treatment T 5 , rice stubble residues were left over intact across the field.
Before tillage operations, soil moisture content was measured.To do that, three spots of each plot were sampled by gimlets to depths of 0-15 cm and 15-30 cm.Samples were placed in special canisters and weighted and then put inside an electrical oven for 24 hours at 105°C.Later, samples were removed from oven and again weighted after reaching ambient temperature inside a desiccator.Soil moisture content was calculated from the following formula (Anonymous, 1995): (1) Where, M c : soil moisture content (%db); W w : weight of soil sample before putting inside oven (g); and W d : weight of dried soil sample (g).
Soil bulk density was measured before and after applying tillage treatments.From three spots in each plot, samples were obtained using core sampling set from middle parts of depth ranges of 0-15 and 15-30 cm.Then, samples were put inside an electrical oven for 24 hours at 105°C and they were weighted again after reaching room temperature in a desiccator.Soil bulk density was computed from the following equation (Anonymous, 1995): (2) Where, ρ d : soil bulk density (gcm -3 ); W s : weight of dried soil (g); and V t : total volume of soil (cm 3 ).
In order to measure plant residues across the field, a wooden frame of 1 m × 1 m was randomly thrown away three times in each plot and all enclosed plant residues were collected and weighted after being dried in an electrical oven for 48 hours at 60°C.The following formula was employed for calculating plant residues (Anonymous, 1995): (3) Where, R s : plant residues after tillage (%); W r : dried mass of surface residues after tillage (g m -2 ); W t : dried mass of surface residues before tillage (g m -2 ).
Crop management was accomplished as instructed by agronomists.Before seed planting, in order to compensate soil nutrition deficit, urea (100 kg ha -1 ) and triple super phosphate (150 kg ha -1 ) fertilizers were added to soil.Urea fertilizer was applied in two stages which rated as 50 kg at the beginning of stem growth and the other 50 kg at the flowering stage in the form of top dressing.The seeding rate of PF variety was selected to be 12 kg ha -1 because of heavy soil texture and high moisture content conditions of paddy fields.
Days required for seed germination in different treatments were recorded.To measure weed density in different methods, a wooden frame (1 m × 1 m) was randomly thrown away three times in each plot roughly 30 days after seeding.Then, weeds enclosed in wooden frame were gathered and weighted after being dried in an electrical oven for 48 hours at 60°C.In order to achieve appropriate plant density per area unit (around 100 plants per m 2 ), thinning operation was done at 4-to 5-leaf stage.
In grain ripening stage, to specify grain yield, an area of 12 m 2 (3m × 4m) was harvested from each plot and grains were weighted after threshing and cleaning and grain yield was computed on the basis of 12% moisture content and in terms of kgha -1 .Data analysis of variance and means comparison were carried out using Duncan's multiple range tests at 5% level by SAS 9 (SAS Institute Inc., Cary, North Carolina, USA).

Soil bulk density
Table 1 illustrates comparison between soil bulk densities of experimental plots in different tillage methods over test years.As it can be seen, in each depth of experimented site, soil bulk densities of tillage treatments (T 1 , T 2 , and T 3 ) were less than the ones of no-till treatments (T 4 and T 5 ) and this was particularly distinguished in the 0-to 15-cm depth.Also, for each tillage treatment, deepening the tillage made considerable increase in soil bulk density.This is especially remarkable in treatments T 1 and T 2 .Working depth of implement would be a reason for this issue.In treatments T 1 and T 2 , where rotavator was used as tillage implement, blades worked narrowly (10-15 cm) and layers lower than 15 cm were not affected by blades.Using moldboard plow for primary tillage (T 3 ) had greater influence on reduction of soil bulk density in depths lower than 15 cm whereas tillage by rotavator (twice) (T 3 ) provided smoother surface for top soil (in 0-to 15-cm depth).Reshad-Sedghi and Zabolestani ( 2001) reported that soil bulk density of top soil was lower when rotavator was employed for tillage operations.Higher soil bulk density in no-till method compared to tillage methods has been noted by other authors (Grant and Lafond, 1993;Beare et al., 1994;Hadj-Abbasi et al., 2000).

Surface residues after tillage
The effect of tillage method on surface residues was significant (p<0.01) for both test years (Table 2).Except T 5 (where all rice stem residues remained intact), the least and most surface residues were associated with T 4 and T 1 (minimum tillage method), respectively.For crop season of 2009-2010, there was no significant difference between T 3 and T 4 in this regard.The results of compound ANOVA indicated that the effect of tillage method on surface residues was significant (p<0.01),but year and interaction of year × tillage method did not affect this trait (Table 4).The highest and lowest surface residues were related to T 1 (64.63%) and T 4 (24.25%),respectively.Surface residue for T 2 was 49.29%, whereas it was 30.08% in T 3 (Table 5).The amount of surface residues depends on the type of tillage implement.When using moldboard plow in primary tillage, roughly 70% of rice stems were buried among soil inverted layers.In plots where rotavator was applied, inversion level of residues was lower than when using moldboard plow so that more than 60% and around 50% of surface stalks remained on soil surface in T 1 and T 2 , respectively.

Seed germination
The results of simple ANOVA showed that the effect of tillage method on days required for seed germination was significant (p<0.05)(Table 2).Means comparison represented that for both test years, seed germination in plots where no-till treatments were applied took more days in comparison with other treatments, i.e.T 1 , T 2 , and T 3 (Table 3).Average days for seed germination in T 4 and T 5 were 8 days for the first test year while it was 7.3 and 7.7 days respectively for the second test year.The shortest time for germination took 6.3 days in average in T 2 and T 3 .The results of compound ANOVA revealed that the effects of year and interaction of tillage method × year were not significant on days for germination whereas tillage method had significant effect (p<0.01) on it (Table 4).The longest time for seed germination was associated with T 5 and T 4 with 7.8 and 7.7 days (mean of two years), respectively but the shortest time was in T 2 with the mean of 6.3 days (Table 5).Seed germination depends on factors such as soil moisture content, the way seed makes contact with soil particles, the level of soil surface coverage by plant residues, and ambient temperature.The reason why seeds germinated in a shorter time in T 2 and T 3 can be attributed to proper contact between seed and soil as a result of plowing and reducing residues on soil surface which led to temperature rising around seed due to absorbing more solar radiation.In T 2 and T 3 , suitable circumstances for seed germination were provided since soil surface became smoother as a result of more plowing operations.In each column, figures sharing a common letter are not significant at 5% level.

Dry mass of weeds
The simple ANOVA implies significant effect of tillage method on dry mass of weeds (Table 2).For crop season of 2009-2010, dry masses of weeds related to T 1 , T 2 , and T 3 were statistically located in the same group whereas T 4 and T 5 had their own group.Though, the least dry mass of weeds was related to T 3 (32.40g m -2 ) and the utmost one was associated withT 4 (74.23 g m -2 ).For crop season of 2010-2011, T 3 had the lowest dry mass of weeds (42.60 g m -2 ), but the highest one was observed in T 5 (71.33 g m -2 ) (Table 3).
The compound ANOVA showed that the effect of tillage method on dry mass of weeds was significant (p<0.01)(Table 4).The effects of year and interaction of year × tillage method were not significant on this trait.Biennial means comparison demonstrated that T 3 had the least dry mass of weeds (37.50 g cm -2 ) while the highest amount pertained to T 5 (71.89 g cm -2 ).In T 3 , by running primary tillage, surface residues and weeds were buried between inverted layers of soil by moldboard plow.Therefore, they did not have enough time to grow again.But, for no-till treatments, i.e.T 4 and T 5 , soil circumstances were in favor of weed growth.Hence, applying efficient ways to control weeds would have crucial role in minimum and no-till methods.

Grain yield
The simple data analysis of variance showed that the effect of tillage methods on grain yield was significant (Table 2).T 2 and T 3 had the maximum grain yield, but the minimum was obtained for T 4 and T 5 (Table 3).The compound data analysis of variance also was an indicative of significant effect of tillage methods on grain yield (Table 4).The effect of year was significant on this trait as well.Some researchers believe in grain yield increase as a result of enhanced plant establishment (Majnoun Hosseini et al., 2006;Inamullah et al., 2013),while others suggest that reducing soil compaction, development and growth of roots inside soil, attracting more nutritious, and aerial organs development are influential in raising grain yield (Leach et al., 1999;Yazdifar and Ramea, 2009;Sarkees, 2013).
It is necessary to underline that the effect of tillage methods on grain yield and yield components of canola in common conditions is different from what it is observed in paddy fields.In an investigation, it was revealed that the highest grain yield was associated with conventional tillage treatment whereas the lowest one was related to no-tillage treatment due to small size of seeds and high sensitivity of rapeseed plant to proper seedbed establishment (Fooladivand et al., 2009).

Table 1 .
Soil bulk density of test site during crop seasons.

Table 2 .
ANOVA for different tillage methods in rapeseed planting after rice.

Table 3 .
Means comparison for different tillage methods in rapeseed planting after rice.
aIn each column, figures sharing a common letter are not significant at 5% level.

Table 4 .
Compound ANOVA for different tillage methods in rapeseed planting after rice.

Table 5 .
Biennial means comparison for tillage methods in canola planting after rice.