Modelagem do fluxo de emergência das plantas daninhas em sistema agroecológico e convencional de produção vegetal

Abstract

The development of weed emergence prediction models helps us to better understand their biology, as well as helps us in the elaboration of the best control strategies. Therefore, the objective of this work was to model the emergence flow of the weed species Conyza canadensis, Leonurus sibiricus and Bidens pilosa, due to agroecological and conventional management. The experiment was conducted from March/2020 to September/2021. Six plots with different crop rotation systems were evaluated, five plots with agroecological management and one plot with conventional management. The systems adopted for each agroecological management were: plot A: wheat/beans/off-season corn + black oats, plot B: oats/soybeans/off-season corn, plot C: Mix 520 (pea + oats + rye + turnip)/corn/wheat , plot D: Mix 110 (Oats + Rye + Turnip)/Soybean/Mix aduverd 1005 (Ry + Vetch + Black Oats + Turnip + Peas), plot E: Mix (Oats + Turnip)/Fallow/Mix aduverd 1005, and for conventional management: Off-season corn/Soybean/Off-season corn (plot F). The counts were performed fortnightly, and the plants that emerged were counted, identified and removed. Each plot had 20 georeferenced points of 0.25 m², 10 sample points with the presence of the crop and 10 sample points without the presence of the crop. During the survey period, 46 weed species were identified. The species Richardia brasiliensis showed the highest IVIr in plots A, E and F; while for plots B, C and E the highest Relative Importance Value Index was obtained by the species Leonurus sibiricus, Bidens subalternans and Parietaria debilis, respectively. The greatest diversity of weed species was found in plot E, with 30 species identified, while the lowest diversity was found in plot F, with 20 species identified. The points without the presence of culture presented more emergency flows and longer flows, in relation to the points with the presence of culture. Crop rotation combined with cover crops reduced the emergence of weeds in agroecological management. The lack of soil cover and fallow increased weed emergence. Among the species present in the areas, it was possible to obtain flow models for Leonurus sibiricus, Bidens subalternans and Conyza. canadensis. The predictability of the Thermal Time and Hydrothermal Time models differed among managements, among sowing times and among growing seasons. The species Leonurus sibiricus and Conyza canadensis showed better fits for the model based on thermal time, while the species Bidens subalternans showed better fits for hydrothermal time. The thermal and hydrothermal model can predict the emergence of Conyza canadensis, Leonurus Sibiricus and Bidens subalternas in agroecological and conventional management.