Deriva e análise de risco das aplicações de fungicidas, herbicidas e inseticidas nas culturas do feijão e da soja

Abstract

The use of pesticides in management of dry beans and soybeans is fundamental to maintain healthy crops. However, inadequate use of these products represent risks of contamination of the fauna, flora and humans through drift during spraying. Thus, this study quantified deposited drift after boom spraying with different nozzles to determine drift curves for dry beans and soybeans and, from such curves, to calculated the indices of environmental and occupational risks of pesticides used for both crops and to establish buffer zones for spraying these products. The study was divided into two parts: the first one estimated drift curves for dry beans and soybean crops, according to the methodology proposed by norm ISO 22866. The experimental design was randomized blocks, in split plots over space 4 x 20 with 10 replications, measuring drift from spraying with four nozzle types (standard and venturi flat fans XR and AIXR and standard and venturi deflector flat fans TT and TTI) in 20 sampling points (corresponding to 20 parallel distances to crop row outside the target area, spaced by 2.5 m). Evaluation of deposition in filter paper targets on the soil, the dye Rodamine B was added to the tank of a hydraulic boom sprayer, which was quantified by fluorimetry. The second part consisted of computations of contamination risk evaluations from exposition to fungicides, herbicides and insecticides recommended for both crops in Brazil. The risk indices were computed for three different groups: water organisms (daphnia, alga and fish), terrestrial (bees and worms) and occupational risk for bystanders, children and residents, according to models proposed by the methodologies POCER and HAIR. Four drift prediction models were obtained for dry beans and three for soybeans, similar to the Dutch Model , with exponential trend of four parameters, considering the 90 percentile. The coefficients of the drift Dutch Model for Cereals, as well as the coefficients of the German Model for Field Crops are not adequate to estimate drift for dry beans and soybeans. In general, water organisms are more sensitive to contamination by drift. No risk of contamination was found for worms. However, the insecticides chlorpyrifos, spinosad and thiamethoxam presented risk to bees, regardless of droplet size used in the drift curve and, therefore, demand different dimensions for the buffer zones. The herbicide paraquat was the only one with potential occupational contamination risk. It is suggested that, for estimation of drift and risk in dry beans and soybeans, for Brazilian climate conditions, the drift prediction models generated in this study, for each nozzle type, should be used.