ALTERNATIVAS DE CONTROLE DE CAPIM AMARGOSO E USO DE PLANTAS DE COBERTURA PARA O MANEJO DE PLANTAS DANINHAS EM ÁREAS DE PLANTIO DIRETO

Abstract

The integration of management strategies is very important in the case of difficult-tocontrol weeds, such as Digitaria insularis. The use of cover crops in the dry season combined with chemical management may reduce problems with herbicide resistant plants. The objective of this work was to evaluate the efficiency of the application of different herbicides for the control of perennial Digitaria insularis, with or without mechanical mowing, and to evaluate the effect of the cultivation of cover crop combinations during off-season on weed suppression in no-till areas. For this, two experiments were conducted; The first one conducted at the Goiano Campus Urutaí IF, Goiás, in a 4x2 factorial scheme, composed of: Glyphosate 1,440 g. e.a. ha-1, Cletodim 108 g i.a. ha-1 and Haloxyfop-p-methyl 62.35 g. i.a. ha-1 and a control without application and two grass management (presence and absence of mowing), in a completely randomized experimental design with four replicates, each experimental unit represented by a vessel with 12 L of capacity. The unriped grass treatments were applied at 120 days after transplanting and the grass was harvested at 90 days after transplanting, with herbicide application after 40 days after mowing, at regrowth. Visual evaluations of phytointoxication were carried out in five seasons until 42 days after herbicide application (DAA), and the aerial part of the plants was collected in the last evaluation period to determine the dry matter. Data were submitted to analysis of variance and Fisher's LSD test for comparisons between means, at 5% probability. The second experiment was conducted at Embrapa Arroz e Feijão, where different combinations of cover crops were cultivated in the no-till cropping season to evaluate weed suppression. The treatments were: Pousio (control), Milheto + Crotalaria, Milheto + Guandu, Milheto + Ruziziensis, Milheto + Ruziziensis + Guandu and Milheto + Buckwheat. Weed identification and determination of infestation density and shoot dry matter production were performed at 30 and 75 days after sowing of cover plants. The total weed infestation and density data of the main species in the area were submitted to deviance analysis from the generalized Poisson and binomial linear model, respectively. The means of total infestation and density treatments of the main species were compared from the 95% confidence intervals with Poisson and binomial distribution, respectively. For the dry matter data of the weeds were submitted to analysis of variance with plots subdivided in time and Fisher's LSD test, at 5% probability. In the control experiment of D. insularis the herbicides did not provide control of the unrooted plants, with phytotoxication below 7.6%. The herbicide application of Glyphosate, Cletodim and Haloxyfop-p-methyl herbicides resulted in a significant control of the grass, with 82%, 59% and 64% phytotoxification, respectively, but without causing the plants to die. showing as feasible alternative the integration of the mechanical control method followed by the chemical control to increase the efficiency in the management sourgrass in areas with high infestation index. For the experiment with cover plants, we observed an effect on the reduction of density and dry matter production of weeds compared to the fallow area.