Assessment of weed control methods for maize production by emerging farmers on commercial farms

Abstract

Maize (Zea mays L) is a staple food crop grown in South Africa by both large scale commercial and smallholder farmers. During the 2013/14 cropping season maize occupied about 2.6 million hectares of the total 3.9 million hectares of arable land that was under field crops in South Africa. Maize accounted for about 12.4 million tonnes of the 14.4 million tonnes of all field crops produced. Excessive competition from weeds is a major constraint, reducing maize yield and farmer income. Resource poor and inexperienced emerging farmers who have acquired land through the government land redistribution programmes are particularly affected. To date about 5.7 million hectares of land have been transferred to about 4.2 million black (previously disadvantaged) emerging farmers. Although emerging farmers have several options available for weed control, these still need to be appraised with regards to benefits, in the form of grain yield measured against the cost of weed management. An on-farm study was carried out at two sites in the North West province of South Africa during the 2011/12 and 2013/14 cropping seasons. The objectives were:- ? To determine the effect of different weed control methods on maize yields of emerging farmers at two localities in the North West province. ? To identify the most competitive or problematic weed species at two localities in the North West province. ? To compare the economic benefit of different weed control methods at the two localities. The experiment was laid out in a split-plot randomised block design. A stacked gene (stalk borer and herbicides resistance) and a conventional maize cultivar were planted in strips. Eight weed control methods that included hand-weeding, mechanical, chemical (herbicides) and combinations of these methods were randomly allocated across the strips. Weed species were counted and crop heights were recorded at three and eight weeks after crop emergence (WACE). Weed dry biomass was also determined. Grain yield and the yield components of ear mass and 100 kernels mass were recorded. A cost-benefit analysis of these weed control methods was carried out in the context of total production costs. The highest maize grain yields were obtained, where weed competitive effects were satisfactorily suppressed. The clean field and pre- and post-emergence herbicides methods produced the highest grain yields in the two seasons. In the first season the highest grain yields obtained were 73% higher than the lowest yield in no-weeding method for both cultivars. The second season was characterized by below average and erratic rainfall. The stacked gene cultivar outperformed the conventional cultivar by 63% where weeds were effectively controlled. Weed competition seemed to cancel the superiority of the stacked gene cultivar over the conventional cultivar in a drier season. The cost-benefit analysis revealed that a single cultivation operation at six WACE was the cheapest method, costing only R 495 ha-1 irrespective of the cultivar used. Keeping a clean field throughout the season was the most costly endeavour, at R 2 528 ha-1 and R 2 174 ha-1 for the conventional and the stacked gene cultivar respectively. The use of both pre- and post-emergence herbicides on stacked gene cultivars can provide farmers with a return of up to R 2.60 for every R 1 invested. Controlling weeds in a conventional maize cultivar, using tractor-drawn cultivator at six WACE, can give a return of up to R 1.64 for every R 1 invested. The weed control methods that provide the highest grain yields are not necessarily the most cost effective. It is preliminarily recommended that chemical weed control methods be considered if stacked gene cultivars are to be planted. However, mechanical weed control methods must be considered when planting conventional cultivars. The present study needs to be intensified, covering a wider geographical extent, to cater for variation that can be expected as a result of differences in climate, soil type and weed spectra.