The adoption of integrated weed management systems depends on the ability to foresee the impact of weeds on crop grain yield potential. The objective of this research was to compare different explicative variables to use in adjusting the rectangular hyperbolic model, in order to identify which one would provide best prediction of beggarticks (Bidens pilosa and Bidens subalternans, occurring in mixed infestations) interference on soybean crop. Experiments were carried out in two environments, located at Passo Fundo and Eldorado do Sul, RS, Brazil. Various beggartick densities and three soybean seeding times (3, 7, and 11 days after cover crops desiccation) were used. Evaluations carried out were beggartick densities 20 and 30 days after soybean emergence (DAE) and at crop harvesting time, and weed plant morphological characteristics, such as leaf density, leaf area, and soil leaf coverage, all performed 20 DAE. It was detected that the explicative variable beggartick plant density provided satisfactory adjustments of the equation model; nevertheless, it does not integrate the effects of environments nor of crop seeding time, differently of what happened with the morphological characteristics. Among them, leaf density and leaf area showed greater potential for use as explicative variables in predicting soybean grain yield losses due to beggartick interference.
Bidens pilosa; Bidens subalternans; competition; modeling; explicative variables
