Effect of row spacing and weed management practices on dry direct-seeded spring rice

Weeds are the major burden for rice-growing farmers in the case of direct-seeded rice due to the preemptive competition of weed in the species early seedling stage. A field experiment was carried out during the spring season in 2021 to evaluate the effect of different weed management practices and row spacing on dry directed-seeded spring rice under the Prime Minister Agriculture Modernization Project (PMAMP) at Rice Super Zone, Kanchanpur, Nepal. The experiment was laid out in a two-factorial randomized complete block design (RCBD) with twelve treatments and three replications. The highest number of effective tillers per m 2 was obtained in weed-free plots (531.67), which is statistically similar to Pretilachlor fb 1HW (505.42). Row spacing of 10 cm showed a significantly higher number of effective tillers per square meter (521.94). A higher grain yield was observed in weed-free plots (6397.87 kg.ha -1 ), which is followed by Pretilachlor fb 1HW, which is statistically similar to other management practices except for weed check. All the management practices were found effective in reducing weed parameters such as weed density, weed dry weight, and weed index as compared to weed checks. A positive relation was recorded between plant height at 90 DAS and grain yield, thousand-grain weight, and grain yield, and a negative relation was recorded between weed density, weed dry biomass, and grain yield. It is recommended to apply Pretilachlor fb 1HW with 10 cm row spacing for effective weed control and to increase the yield of dry direct-seeded rice.


Introduction dios de esspinoza
Rice (Oryzae sativa) is the third major cereal crop in the world and the first major crop in Nepal in terms of production (Agriculture Information and Communication Centre [AICC], 2019).It is a major source of calories for over half of the world's population (Kumar & Ladha, 2011).According to Ministry of Agriculture and Livestock Development (MoALD, 2022), it has contributed 20 % to agriculture gross domestic product (AGDP) and 7 % to gross domestic product (GDP).It is cultivated in Nepal at an altitude of 3 050 masl, from Jumla to the lowest point of 59 masl, Kachankawal (Paudel, 2011).Kanchanpur district has a latitude of 28° 50'13.9200"north and a longitude of 80°19'16.7232east with an altitude of 176 m to 152 m from the masl.The total area cultivated under rice in Nepal was 1 458 915 ha, and the production was 5 550 878 t in the fiscal year 2077/78, with a productivity of 3.80 t.ha -1 .Similarly, the total area under rice in Kanchanpur district under main season and spring rice was 47 686 ha and 789 ha respectively, with a total production of 185 191 and 35.66 t, giving productivity of 3.88 t.ha -1 and 5.12 t.ha -1 , respectively (MoALD, 2022).
Rice is predominantly cultivated by transplanting in puddled soil with continuous flooding.Puddled transplantation causes the formation of hard pan at the sub-surface layer, thereby reducing soil permeability.Also, it is found that this method affects soil physical properties, affecting the growth and productivity of rice and succeeding wheat crops too (Gathala et al., 2011).Generally, transplanted rice suppressed the population of weeds through the formation of an anaerobic environment and reduced the percolation loss of water.But this method requires a lot of water and labor for land preparation, nursery establishment, and transplanting.Zhao et al (2007) reported that dry direct-seeded rice (DDSR) requires 25 % -30 % less water as compared to puddled-transplanted rice.In contrast to transplanted rice, directseeded rice as an alternative is cost-effective and requires less labor.Labor used in direct fields is about 50 % lower than in puddled transplanted fields Kumar & Ladha (2011).The adoption of the direct-seeded method resulted in a 5.33 % increase in grain yield and a 7.51 % reduction in total production costs (Zhao et al., 2007;Mishra et al., 2017)the availability of appropriate land equipment, and irrigation-drainage systems.Using plot-and household-level data, we analyze the impacts of DSR adoption in two rice-growing states of India.We account for observed and unobserved heterogeneity using endogenous switching regression.We analyze the yield and costs effects of DSR adoption.Our study shows a small but significant effect of DSR adoption on yield and costs.We find increase in rice yields (by 3.74 %).
Weed management is a major constraint in direct-seeded rice (Ranjit & Suwanketnikom, 2005).The competition occurs between weeds and rice for sunlight, nutrients, space, and water, thereby reducing the potential yield.The order of weed dominance was 58 % broadleaf weeds, 20 % sedges, and 17 % types of grass at the DRR farm, which varies with location (Subbaiah, 2008).Furthermore, the random use of herbicides in the name of weed management has had brought several negative impacts on the environment.It causes harm to soil microorganisms, deteriorates the chemical properties of soil, and aids in the development of resistance in weeds against herbicides.Ayansina & Oso (2006) reported that herbicide treatments, even at the recommended dose, resulted in a reduced soil microbial count.Thus, row spacing can be an option for integrated weed management in combination with herbicides.Herbicides such as Pretilachlor and Pendimethalin were found effective for weed management in dry-direct seeded rice (Jabran & Chauhan, 2015).The morpho-physiological parameters such as plant height, tiller number pre-hill, total dry matter per hill, and yield of the plant are greatly affected by row spacing in rice (Roy et al., 2014).Thus, this research aims to evaluate the effect of weed management practices and row spacing on the performance of DDSR during the spring season.

Physio-chemical properties of the experimental soil
Soil samples were collected randomly from every four corners and a center experimental plot at 0 cm -20 cm depth from the surface using a shovel to analyze the initial physio-chemical properties.From the analysis, the soil texture of the research plot was found to be sandy loam.The pH of the soil was neutral (6.84) in nature.The total nitrogen content of the soil was low (0.08 %), while available phosphorus (20.06 kg.ha -1 ) and available potassium (121.2 kg.ha -1 ) were low and medium, respectively.The organic matter content of the soil was low (1.68 %).

Weather conditions during the experiment
The experiment site lies in the sub-tropical region of Nepal.During the experimental period, the maximum temperature was 25 °C in June, while the minimum temperature was 14 °C in February.The relative humidity ranges from 29 % in April to 50 % in June as shown in Figure 1.

Sample and sampling techniques
First, the preliminary survey for major problems related to direct seeded rice was carried out near the areas of the superzone.The field research for weed management under non-irrigated land by DSR (direct seeded rice) was conducted in the field of farmers who have been cultivating directseeded rice for a few years.The samples were collected from the replications made on the plots of rice fields.The samples from each plot were taken randomly excluding the border rows.

Experimental designs
The experimental design was two factorial RCBD with 12 treatments (Table 1) and 3 replications.Individual plot size was 2 m 2 ×2 m 2 and the row spacing was 10 cm and 20 cm.The spacing between the blocks was 1m and the spacing between treatments was 0.5 m.

Crop management practices Fertilizer application:
N, P, and K were applied at the recommended dose in the ratio of 120:40:40 kg.ha -1 , respectively.The full dose of potassium and phosphorus was applied as a basal.While one-third of the nitrogen was applied as basal and the remaining was applied in two split doses as top dressing at tillering and panicle initiation stages (Dangol et al., 2020).

Seed rate and sowing:
The sowing of seeds was done on 24 th February 2021.They were sown directly in rows by maintaining the spacing of 10 cm and 20 cm.The seed rate was 24 g/plot.The seeds were first treated with salt at a rate of 200-300 g/L of water (Dangol et al., 2020).The seeds were sown manually.

Treatment Application and Inter-cultural Operations:
One Hand weeding was practiced at 30 DAS (days after sowing), and in weed-free plots, hand weeding was practiced at intervals of 7 days up to the panicle initiation stage.Gap filling was done at 15 DAS.Imidacloprid was sprayed at 20 DAS to control aphids and plant hopper (Table 2).

Measurement 2.6.1 Data related to Weeds Weed density:
The number of weeds that emerged was counted in the area between the 18 th and 19 th row and 8 th and 9 th rows of rice in 10cm and 20cm row spacing respectively, i.e., from 0.1 m 2 (50 cm × 20 cm) at 30, 60, and 90 DAS.

Weed dry weight:
Weed samples were taken from the inter-row space between the 18 th and 19 th and 8 th and 9 th intervals in row spacing of 10 cm and 20 cm, respectively, at 30 DAS, 60 DAS, and 90 DAS for each treatment.At the time of sampling, weeds were taken from an area of 0.1m 2 (50cm × 20cm) and values were converted to per square meter.Dry matter deposition was determined by oven-drying weed samples at 70 °C for 48 hours.Weed Index (WI): It refers to the reduction in yield due to the presence of weeds in comparison to the weedfree plot.It was calculated using the following formula given by Gill & Vijaya Kumar (1969).

WI= [(X-Y)/X]* 100
Where, X= Yield from the weed-free plot Y= Yield from the treated plot

Biometrical observations
Plant height (cm): Plant height was measured from randomly selected plants excluding border rows at an interval of 15 days from 30 DAS up to 90 DAS.For height, the vertical distance was measured from the ground level of the plant to the longest leaf of the main tiller with the help of scale and was averaged.
The number of tillers/m 2 : It was calculated by using a non-destructive sampling technique at 15-day intervals from 30 days after sowing up to 90 days after sowing and the average was computed.

Yield and yield attributing parameters of Rice
The number of effective tillers/m 2 : An effective tiller refers to that tiller which yields panicle on it.It was computed just before the harvesting stage.
Length of panicle, grain/panicle, and sterility percentage They were calculated from randomly selected 20 panicles excluding border rows just before harvesting.The sterility percentage was computed by determining the number of filled and unfilled florets from the selected panicles.
The sterility percentage was computed by using the following formula, Sterility (%)=[number of sterile florets/total number of florets] × 100

Thousand-grain weight:
It was computed from the randomly selected grains, and weighed by using an electronic balance.

Grain and straw yield (kg ha -1 ):
Moisture content in grains was measured by using Wile 65 moisture meter.Grain yield at 14 % moisture was calculated using the formula suggested by Paudel (1995).For straw yield, the grain yield was deducted from the total biomass. where,

Harvest index:
It is the ratio of the economic yield to the total yield of biomass.It was computed by using the given formula, Harvest index (HI) = (Grain yield )/(Total biomass yield)

Results and Discussion
3.1 Influence of weed management practices and row spacing on crop parameters 3.1.1Biometrical data 3.1.1.1Plant height Plant height in dry DSR was significantly influenced by weed management practices, but on row spacings of 10 cm and 20 cm, the plant height was found to be statistically similar during all days of observation from 30 DAS to 90 DAS (Table 3).Row spacing had no significant influence on plant height (Payman & Singh, 2008).At all dates of observations, the tallest plant height was recorded in weed-free plots, whereas the shortest plant height was observed on weedy check plots.At 30 DAS, the plant Grain yield (kg.ha -1 ) at 14 % moisture = ((100-MC)×plot yield (kg)×10000m 2 )/((100-14)×netplot area (in square meter))  (327,739,896,994,544) as compared to 20 cm row spacing (203,465,697,840), respectively (Table 4).The number of tillers increases with increasing row spacing, which is similar to the finding of Roy et al (2014).But, in the case of unit area basis, the number of tillers in narrow row spacing was found to be higher at a row spacing of 10 cm.Note: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.
Treatment mean followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05).

Effective tillers/m 2
The effective tillers per square meter of dry DSR were significantly influenced by weed management practices and row spacing.Weedfree plots had the highest number of effective tillers per square meter numerically (531.67), which was statistically superior to other treatments but statistically at par with pretilachlor fb 1HW.
The number of effective tillers per meter square was found to be significantly lower in weedy check plots numerically (332.50) because of higher weed density and dry weight, along with a lower number of tillers per meter square.The higher number of effective tillers in weed-free and pretilachlor fb 1HW-treated plots might be due to the higher number of tillers.The effective tillers per square meter were statistically higher at a row spacing of 10cm as compared with a row spacing of 20 cm.In a wider row spacing of 20 cm, plants get more nutrients, moisture, and light, as a result of which the number of effective tillers becomes higher, but per unit area, it decreases due to the lesser number of rows.

Panicle length, sterility percentage, and thousand-grain weight
The weed management practices had a significant influence on panicle length but not on the sterility percentage and thousand-grain weight.The pretilachlor fb 2,4-d EE-treated plots had the longest panicle (24.50 cm) among all the treatments and were statistically similar to pendimethalin fb 2,4-d EE (Table 5).The panicle length was observed to be statistically shorter in weedy check plots, which might be due to nutrient drain resulting from higher cropweed competition for limited nutrients.The row spacing had no significant influence on sterility percentage, panicle length, and thousand-grain weight.Probably, the parameter thousand grain weight, is not affected by row spacing because it is a varietal character and controlled by the hull (Yadanar et al., 2018).

Yield 3.1.3.1 Grain Yield, straw yield, and harvest index
The different weed management practices were found to have a significant influence on grain yield and straw yield, but not on harvest index.The highest grain yield was recorded on the weed-free plots (6394.87kg ha -1 ).Pretilachlor fb 1HW, pendimethalin fb 1HW, and pretilachlor fb 2,4-D EE-treated plots produced significantly higher grain yield than the weedy check.Pendimethalin fb 2,4-D EE plots were produced statistically at par grain yield to weedy check plot (Table 6).
The statistically lower position was obtained by Weedy check plots (2119.19kg ha -1 ), which might be due to inferior performance in terms of some yield-attributing parameters.Increased grain yield as a result of an effective weedy check through herbicides was reported by Ishaya et al (2007).Akbar et al (2011) and Choudhary & Dixit (2018) reported that the order of herbicides in improving grain yield was pretilachlor, followed by pendimethalin.The grain yield of direct-seeded spring rice was maximum on plots treated with pretilachlor at a rate of 900 g ai/ha (Chauhan et al., 2014).The highest dry straw yield was 16943.14kg ha -1 recorded with weedfree plots, which were found to be significantly similar to pretilachlor fb 2,4-D EE-treated plots.
Among the six weed treatments, pretilachlor fb 1HW-treated plots occupied statistically similar positions to other weed-treated plots.The row spacing had no significant impact on grain yield, straw yield, and harvest index.Chauhan & Johnson (2011) also reported no significant impact of row spacing between 15 cm and 10-20-10 cm rows on weed growth and yield of aerobic rice.

Influence of weed management practices and row spacing on weed parameters 3.2.1 Weed density
The weed density was significantly influenced by weed management practices except at 90 DAS and also had no significant influence by row spacing except at 60 DAS (Table 7).The integration of narrow row spacing with a high seed rate helps in effectively controlling the weed population (Matloob et al., 2015).At 30 DAS, the significantly highest weed density was observed in the weedy check, and the lowest was observed in pendimethalin fb 1HW.Mishra & Singh (2008) also reported significant control of Echinochloa colona on the application of pendimethalin 1 kg/ha fb 1 HW in DDSR.At 30 DAS, the pendimethalin fb 1HW was statistically similar to the remaining treated plot except the weed-free plot.At 60 DAS, the significantly lowest weed density was obtained in the weedfree plots, which was at par with pendimethalin fb 2,4-d EE.The effect of pendimethalin fb 2,4-d EE and pretilachlor fb 2,4-d EE on weed density was observed to be similar at 60 DAS.
Chauhan et al ( 2014) also reported the lowest weed density at a high dose of pretilachlor in direct-seeded spring rice.

Weed dry weight (g/m 2 )
The dry weight of weed biomass was significantly influenced by weed management practices but not significantly influenced by row spacing (F-test, p<0.01).At 30 DAS, the weed-free plot is statistically similar to the remaining treated plot.At 60 DAS, weed biomass in pendimethalin fb 2,4-D EE, pendimethalin 1HW, pretilachlor fb 2,4-D EE, and pretilachlor 1HW is statistically similar to each other.Chauhan et al (2014) found significantly low weed biomass with an increase in the dose of pretilachlor.The application of pretilachlor at a rate of 0.5 kg/ ha resulted in decreased weed biomass and increased grain yield in direct-seeded rice (Phukan et al., 2021).At 90 DAS weedy dry matter, the weedy check plot was found to be similar to the pendimethalin 2,4-D EE and pretilachlor fb 2,4-D EE-treated plots (Table 8).

Weed index
The weed index of DDSR as influenced by different weed management practices is presented in Figure 2. The grand mean weed index for the experiment was 31.57% and ranges from 23.67 % to 67.85 %.The weed index was significantly influenced by various weed management practices.Among the weed management practices, there was a 66.78 % yield reduction in the weedy check plot, which was significantly higher and statistically similar to pendimethalin fb 2,4-D EE.The weed index of

Regression analysis of various parameters 3.3.1 Plant height at 90 DAS and grain yield
There is a linear relationship between Plant height at 90 DAS and grain yield.The Plant height at 90 DAS had only a 67.6 % contribution to the total grain yield and the remaining contribution was due to other factors (Figure 3).

Weed Density and grain yield
There is a linear relationship between weed density at 30 DAS and 60 DAS with grain yield.The weed density at 30 DAS had only a 55.1 % reduction to the total grain yield and the remaining contribution was due to other factors.The weed density at 60 DAS had only a 50.6 % reduction to the total grain yield and the remaining contribution was due to other factors (Figure 4).

Conclusion
The different weed management practices had a significant influence on the growth and yield parameters of DDSR and weeds.After weedfree plots, pendimethalin fb 1HW showed significantly more plant height, which was at par with pretilachlor fb 1HW.The yield attributing parameter, effective tillers/m 2 , was significantly highest in weed-free plots, which were at par with pretilachlor fb 1HW.The panicle length was highest in plots treated with pretilachlor fb 2,4-d EE, which was similar to pendimethalin fb 2,4-d EE.The weed-free plots showed maximum grain yield, followed by pretilachlor fb 1HW, which was similar to all the other management practices except weed check.However, weed check and pendimethalin fb 2,4-D EE showed similar effects on grain yield.After weed-free plots, all the weed management practices except weedy check were at par in controlling weed density, weed dry weight, and weed index.The weed index observed in the weedy check and pendimethalin fb 2,4-d EE were statistically similar.There was a reduction in yield of 67.85 % due to the presence of weeds as compared to weed-free.Among row spacing, tiller and effective tiller per square meter were found to be higher in 10 cm row spacing as compared with 20 cm row spacing.During the overall date of observation, there was no significant influence of row spacing on weed density and weed dry weight except weed density at 60 DAS.The treatment with pretilachlor fb 1HW and 10 cm row spacing is recommended for improving the growth, yield, and yield attributing parameters of dry-direct seeded rice and also for effective control of the weed population.

Figure 1 .
Figure 1.Weather conditions of the research site from January to June 2021

Figure 2 :
Figure 2: Weed index (%) as influenced by different weed management practices

Figure 3 :
Figure 3: Relation between plant height at 90 DAS and grain yield as influenced by weed management practices at Kanchanpur, Nepal, 2021

Figure 4 :
Figure 4: Relation between weed density at 30 DAS and 60 DAS and grain yield as influenced by weed management practices at Kanchanpur, Nepal, 2021

Table 3 . Influence of weed management practices and row spacing on plant height (cm) of dry DSR at different dates of observation at Kanchanpur, Nepal, 2021
At 30DAS, 45DAS, and 60 DAS, the highest number of tillers per square meter was observed on the weed-free plot, followed by the pretilachlor fb 1HW-treated plot and pendimethalin fb 1HW-treated plot.At 30 DAS, the number of tillers per square meter on pendimethalin fb 1HW was found to be statistically similar with pendimethalin fb 2,4-D EE and Pretilachlor fb 2,4-D EE-treated plots.At 45 DAS, the numbers of tillers per square meter in pendimethalin fb 2,4-D EE were found to be at par to pretilachlor fb 2,4-D EE and weedy check.At 60 DAS and 75 DAS, pendimethalin fb 1HW and pretilachlor fb 1HW treated plots were found to be statistically similar to the weedy check plot.At 90 DAS, numbers of tillers per square meter in weed-free plots and pretilachlor fb 1HW were found statistically superior, followed by pendimethalin fb 1HW and pretilachlor fb 2,4-d EE respectively.At 90 DAS, the number of tillers per square meter in the weedy check plot is the lowest (380.00).At all dates of observation, the highest number of tillers was statistically observed in 10cm row spacing Note: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.Treatment means followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05).3.1.1.2Number of tillers/m 2

Table 5 . Influence of weed management practices and row spacing on yield attributes of dry DSR at different dates of observation at Kanchanpur, Nepal, 2021
Note: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.Treatment mean followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05).

Table 6 . Influence of weed management practices and row spacing on yield per hectare of dry DSR at different dates of observation at Kanchanpur, Nepal, 2021
: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.Treatment means followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05). Note

Table 7 . Influence of weed management practices and row spacing on Total weed density at DSR at different dates of observation at Kanchanpur, Nepal, 2021
Note: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.Treatment mean followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05).The figures in the parenthesis represent the original value and outside the parenthesis the square root transformation value(√(x+0.5)

Table 8 . Influence of weed management practices and row spacing on weed dry weight at DSR at different dates of observation at Kanchanpur, Nepal, 2021
Note: fb, followed by; DAS, days after sowing; SEm, standard error of the mean; LSD, least significance difference, CV, coefficient of variation.Treatment mean followed by the same letter(s) are non-significance differences on Duncan multiple range test (p > 0.05); the numbers in parenthesis represent the original value and outside the parenthesis the square root transformation value(√(x+0.5)