Integrated Hand Weeding and Herbicide Application for Corn Marigold (Glebionis segetum) in Tef Fields across Western and Southwestern Shewa Zones in Ethiopia

Ethics

This experiment was conducted within an appropriate ethical framework of Ethiopian Institute of Agricultural Research as confirmed by institutional letter with Ref. No.: 9.7/4712/2023 written on 3^rd March 2023.

Study area

The study was conducted at Cheliya and Woliso districts in Oromia regional state in farmers’ fields for consecutive two years of 2020 and 2021 main cropping season (Figure 3). A naturally corn marigold infested field per district was selected (Table 1 and Figure 3).

Figure 3. Map of Ethiopia showing regions, zones and districts at which the experiments were conducted.

Treatments, experimental design and crop husbandry

The experiment consisted of six treatments arranged in randomized complete block design with three replications on the plot size of 4 m × 5 m. Spacing 1.5 m between blocks and 1m between plots were used. The Kuncho variety of tef was drilled in rows in rows 20 cm apart at rate of 10 kg ha^-1. NPS (compound fertilizer containing highly uniform granules of three important plant nutrients, nitrogen, phosphate, and sulfur, with a ratio of 19% N, 38% P2O5, and 7% S) and urea (CO(NH2)2) with 46% N fertilizers (Tufa et al., 2022) were applied at rates of 150 and 100 kg ha^-1, respectively (Yared et al., 2020). NPS was applied at sowing time, whereas urea was applied in two split half amounts at sowing time and the remaining amount at the tillering stage of the crop. The herbicide amount required for the treatment was calculated and measured using a sensitive digital balance and measuring cylinder and applied 30 days after sowing (DAS) Table 2. For all herbicide treatments, a uniform spray volume of 200 L ha^-1 was used. A knapsack sprayer of 20 L volume with three nozzles was used. Moreover, all experimental plots management activities were executed to the standard level.

Data collected

Both corn marigold weed and tef crop data were recorded. The weed parameters such as the number of corn marigold before and after treatment, biomass and visual scoring scale were recorded using appropriate materials and procedures. The weed control efficacy percentage was calculated using the appropriate formula. Tef parameters plant height, biomass, and grain yield were recorded, and a harvest index was calculated.

Number of corn marigold: was counted at 30^th and 60^th days after sowing using 1 m * 1 m quadrat.

Visual scoring: was undertaken using 1-5 scale scoring system at 15^th and 30^th, days after herbicide application (Taye et al., 2007).

Biomass of corn marigold (kg/plot): the weed shoot biomass at crop physiological maturity from the net plot area was recorded and weed control efficacy was calculated using the following formula (Boutagayout et al., 2023a):

Where, WDC and WDT are weed dry weight in control and any particular treatment, respectively.

Plant height: was measured from the ground level to the tip of ten randomly selected plants from harvestable rows at physiological maturity.

Grain yield (kg ha^-1) was measured by harvesting the crop from the net plot area.

Shoot biomass yield: At physiological maturity, the shoot biomass of ten plants selected randomly from the destructive rows was measured after oven drying the harvested produce till a constant weight (Boutagayout et al., 2023b).

Harvest index: The plants harvested for shoot biomass yield at physiological maturity was used to calculate the harvest index by dividing grain yield by the total shoot dry biomass yield.

Relative tef yield loss due to weed was calculated as follows:

Where, MY = maximum yield from a treatment, YT = yield from particular treatment

Where, CFy = crop yield in corn marigold free plots, FPy = crop yield in farmers’ practice, and CCy = crop yield in corn marigold un-weeded check plot.

Data on inputs and their costs were collected for each evaluated weed control option/treatment, including the costs of procuring herbicides and planting materials, labor expenses for land preparation, tillage and herbicide application, hand weeding, and tef harvesting.

The costs of Pyroxsulam 45g L^-1 (2974-birr L^-1), Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 (10913-birr kg^-1), and Flurasulam 75 g L^-1 + Flumetsulam 100 g L^-1 175 SC (8400-birr kg^-1) were obtained from the prevailing local market (53.37 Ethiopian Birr = $1 United State during merchandising). The shopping unit value of the knapsack sprayer was 1200 birr as information gathered from Addis Ababa (central market), Ethiopia. Around West and Southwest Shewa zones, the labor cost man^-1 day^-1 was ranges from 50-150 birr. In addition, data on the market price of tef per kilogram were collected at each site. The current market price (as of January 2023) was taken from farmers who sell tef in local markets and district extension workers to compute benefit cost ratio.

Data analysis

Data that included weed dry weight, weed control efficacy, Tef plant height, above ground biomass, grain yield and harvest index were subjected to analysis of variance (ANOVA) using the SAS statistical software version 9.4 packages (Westfall et al., 2011). The assumption of (ANOVA) normality test was executed for all collected parameters before analyzing the data. Mean separation was performed using Least Significance Difference (LSD) test at 5%.

Cost-benefit analysis

The cost-benefit analysis for the proposed management practices, which included herbicidal and cultural (manual weeding) strategies, was determined using the (Program et al., 1988) process. During the cost-benefit analysis, the total input cost of production, gross revenue, marginal revenue and benefit-cost ratio were all taken into account. The total input cost (additional expenses for weed and trial management) was calculated by adding all costs (variable + fixed input costs) incurred during the study. Tillage and fertilizer costs, as well as planting and harvesting wages, were considered fixed production costs. While the knapsack sprayer, herbicides, and manpower for pesticide spraying and manual weeding were considered variable production costs.

The gross revenue was concluded by multiplying of commercialized price and grain yield (Daramola et al., 2019). The marginal revenue was computed as subtracting the total variable costs from the gross revenue (Daramola et al., 2019) In addition, the benefit cost ratio was computed as the proportion of marginal revenue (numerator) and total variable cost (denominator) (Daramola et al., 2019).

Analysis of variance

The variance analysis revealed statistically significant differences in the evaluated weed management strategies on both the corn marigold and tef characteristics (Tables 3–6). Different corn marigold management approaches produced different results in weed parameters including weed biomass and weed control efficiency, as well as crop factors like plant height, biomass yield, grain yield, and harvest index.

Weeds dry weight

There was statistically significant difference among management strategies in reducing corn marigold dry weight across the study locations (Tables 3 and 4). In the weed-free plots, there was no weed dry weight recorded at the study sites. At Woliso the application of Pyroxsulam 45 g L^-1 supplemented with once-hand weeding significantly decreased the weed dry weight by about 41-fold following the weed free plots. Following the application of Pyroxsulam 45 g L^-1 supplemented with once-hand weeding, the application of Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 supplemented with once-hand weeding significantly decreased the weed dry weight by about 29-fold as compared to weedy check plots at Woliso.

Similar trends were found between the two research locations (Woliso and Cheliya). At Cheliya the application of Pyroxsulam 45 g L^-1 supplemented with once-hand weeding significantly decreased the weed dry weight by about 28-fold following the weed free plots. Following the application of Pyroxsulam 45 g L^-1 supplemented with once-hand weeding, the application of Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 supplemented with once-hand weeding significantly decreased the weed dry weight by about 26-fold as compared to weedy check plots at Cheliya. Due to the higher density of the weed at Cheliya than at Woliso, the reduction in dry weight of the weed there was less than that of at Woliso.

Weed control efficiency

There was statistically significant difference among management strategies in corn marigold control efficiency across the study locations (Tables 3 and 4). Continuous hand weeding of corn marigold as it appeared in the field resulted in complete eradication, whereas Pyroxsulam 45 g L^-1 supplemented with once hand weeding followed by Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 supplemented with once hand weeding resulted in effective corn marigold destruction across the location in similar trend. However, frequent hand-weeding of the weed in tef field is one of the most difficult tasks faced by farmers, especially during the main cropping season. It requires 3–4 rounds of continued weed removal per season to make the field weed-free. Uprooting the weed is difficult due to its complex root system, and cutting it worsens its infestation. Additionally, Tef seedlings are also uprooted and the crop population reduced during frequent hand weeding.

Yield and yields attributes

The effects of weed management practices on Tef plant height, biomass yield, grain yield, and harvest index were significantly different among the management practices across the study locations (5 and 6). However, in terms of plant height, above-ground biomass, and grain yield, there was a statistically insignificant difference among weed-free plot, Pyroxsulam 45 g L^-1 spray supplemented with one hand weeding, and Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 spray supplemented with one hand weeding (5 and 6) across the study locations.

At Cheliya weed-free plots, Pyroxsulam 45 g L^-1 spray supplemented with one hand weeding and Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 spray supplemented with one hand weeding increased tef grain yield by 170, 170, and 126%, respectively. Whereas at Woliso weed-free plots, Pyroxsulam 45 g L^-1 spray supplemented with one hand weeding and Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 spray supplemented with one hand weeding increased Tef grain yield by 31, 30, and 18%, respectively. Overall, the results showed good agreement in the trends of all yield-related parameters across the study locations.

Effect of corn marigold weed on yield loss in tef

The highest yield was obtained from weed free plots, followed by the application of Pyroxsulam 45 g L^-1 at 0.40 L ha^-1 with supplementary hand weeding. The minimal relative yield loss 0.00% at Cheliya and 0.80% at Woliso due to corn marigold was obtained from application of Pyroxsulam 45 g L^-1 at 0.40 L ha^-1 with supplementary hand weeding, followed by application of Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 at 50 g ha^-1 with one supplementary hand weeding with intermediate relative yield loss of 16.28% at Cheliya and 9.99% at Woliso (Table 7).

Based on corn marigold density per area, the potential tef yield loss by the weed ranged from 23.75 to 62.93%, whereas the actual tef yield loss was 13.67 to 47.09% (Table 8). Both actual and potential yield losses due to the weed were highest at Cheliya District, whereas yield losses due to the weed were lowest at Woliso District. This might be due to the fact that the weed density in Cheliya District was higher than that in Woliso District.

Cost-benefit analysis

For the integrated use of herbicide application and supplementary hand weeding for corn marigold management, the marginal revenue (MR) and benefit-cost ratio (BCR) were computed for treatment combinations. The cost-benefit analysis showed that significant variation in MR and BCR was observed among the evaluated experimental treatments (Table 9). Comparing the management practices, the most prominent MR 72459- and 71210-birr ha^-1) and BCR (2.86 and 2.63) were calculated from application of Pyroxsulam 45 g L^-1 supplemented with hand weeding and weed free plots, respectively. Conversely, the lowest MR (34633- and 46571-birr ha^-1) and BCR (1.51 and 1.97) were computed from weedy check and application of Flurasulam 75 g L^-1 + Flumetsulam 100 g L^-1 supplemented with hand weeding, respectively, whereas the intermediate MR and BCR were obtained from the application of Halauxifen-methyl 100 g kg^-1 + Florasulam 100 g kg^-1 with supplementary hand weeding. The MR and BCR obtained from the marketing of goods for application of Pyroxsulam 45 g L^-1 supplemented with hand weeding of corn marigold increased by 110 and 89% as compared to corn marigold weedy check (control plot) (Table 9).

Cost-benefit analysis indicated that integration of Pyroxsulam 45 g L^-1 with supplementary hand weeding exhibited the topmost MR 72459 Birr ha^-1) and BCR (2.86), followed by the weed free plots correspondingly with the MR of 71210-birr ha^-1 and BCR of 2.63. Overall, integrated use of broad leaf killer herbicides with supplementary hand weeding provides a better MR and BCR than single weed management practices (Table 9).