Assessment of the agronomic performance of maize (Zea mays L.) under different weed management regimes

Abstract

Weed management constitutes a major operation and cost in maize production. Three field experiments were conducted in the arable cropping fields of Jomo Kenyatta University of Agriculture and Technology (JKUAT), Juja, Kenya from August 2015 to August 2016 to assess the composition and diversity of weed plants, determine the effect of different weed management practices on weed growth and agronomic performance of maize, and identify the most cost effective weed management option in arable cropping fields. In order to assess the composition and diversity of weed plants, weeds were sampled for two seasons during the fallow phase from three arable cropping fields each measuring about 4 ha and having different cropping histories. Each field was stratified into equal sized plots from which 3 plots were surveyed for weeds using 5 randomly distributed quadrats measuring 0.5m x 0.5m. All weeds within each quadrat were counted and segregated into different species. Weed plants composition was determined by relative abundance, relative density and relative frequency between fields. A total of 28 weed plant species representing 11 families were recorded. Broad leaved weeds (18) were more abundant than grasses (8) and sedges (2).Shannon index for weed diversity was higher in crop research field(Wet season 2.427; Dry season 2.524) compared to maize (Wet season 2.381; Dry season 2.118)and vegetable fields (Wet season 2.05; Dry season 1.862).Wet season increased species richness than dry season by 20% in Crop research field, 36% in Maize field and 21% in Vegetable field. Kruskal-wallis H test results indicate that there was no significant effect of arable cropping system on mean abundance of weeds, H (2) = 2.263, P= 0.322. However, there was high significant effect of season on weed mean abundances, H (1) = 14.173, P= 0.000. To determine the effect of different weed management treatments on weed growth and agronomic performance of maize and to identify the most cost effective weed management option, twelve weed management treatments involving a combination of hand weeding, pre-emergence and/or post-emergence herbicides were applied at 1, 25 and 50 days after planting (DAP); including a weedy check. Treatments were laid out in a RCBD and replicated thrice. Data on dry matter of weeds from different plots was used to calculate weed control efficiency (WCE) and data on maize grain, stover yield, cost of inputs and sale income used to perform gross margin analysis. Results indicated no significant difference for agronomic traits in 2015 growth season. In the 2016 season however, weed density, weed dry weight, cob length and kernels/row were significantly different among treatments. T1 (weedy check) recorded the highest weed density (193.33 g/m2) and T12 (post emergence herbicide at 25 and 50 DAP) the least (6.4 g/m2). Thus, the lowest and highest WCE were recorded in T1 (0%) and T12 (98.1%), respectively. Combined analysis indicated significant seasonal effect for all agronomic traits studied except number of kernels/ row and 100 kernels weight. T6 (hand weeding and post emergence herbicide application at 25 and 50 DAP respectively) produced the highest grain yield (3373Kg/ha) and T1 the least (2394 Kg/ha). The 2016 season gave higher mean values than 2015 for traits studied. On both grain and stover yield, T10 (post-emergence herbicide application at 25 DAP) had highest gross margin (Ksh124, 407) (1244.07 US$) and T5 (two hand weedings at 25 and 50 DAP) lowest (Ksh64, 837.60) (648.4 US$). The results further showed that T6was the most effective weed management option while T10 was the most cost effective treatment.T10 gave the highest gross margin and achieved fairly good grain and Stover yield. Therefore, on the basis of this study T10 may be used by the smallholder farmers for controlling weeds in maize with fairly good yield and economic returns.