Deriva simulada do herbicida dicamba em plantas de Coffea arabica L.

Abstract

Brazil stands out in coffee production, being the largest producer and exporter of the product. This crop, of great importance for the economy, may suffer interference from adjacent crops throughout its entire cycle, from the beginning of cultivation to the stage of full productive development. These interferences can be as diverse as possible, highlighting the drift of pesticides, mainly the drift of non-selective herbicides to the coffee tree. Since the pesticide application process is routine, drift is a factor to be considered in every process. Among the most diverse herbicides used in agriculture, dicamba and glyphosate are highlighted, which are often used in joint applications. However, one of the challenges of using dicamba is the phytotoxicity caused in crops not targeted by the application in a situation of drift, either by volatilization of the active, or by direct drift of sprayed drops. There are several studies evaluating the phytotoxic effects of this active on the most diverse plant species, but there are still few studies focused on coffee trees, in a situation of dicamba drift. Thus, the aim of this study was to understand the phytotoxic effects caused by the simulated drift of the herbicide dicamba alone and in a mixture with glyphosate on the coffee crop (Coffea arabica L.), cultivar Mundo Novo, correlating them with effects on productivity and biometric characteristics in plants in the productive stage and in the initial stage of development. For this, two field experiments were carried out, the first evaluating the visual, biometric, and productive phytotoxic parameters in a commercial area in production. The second was carried out seeking to understand the effects of the treatments used on plants at an early stage of development, evaluating visual phytotoxic parameters and biometric characteristics. Both experiments were conducted in duplicate (experiment I and experiment II), in a randomized block design, in a 2 × 5 + 1 factorial scheme, with 4 replications, where two herbicide mixture compositions (dicamba and dicamba + glyphosate) were considered and five subdoses (0.25; 1; 5; 10 and 20%) based on the full dose of 560 g a.e. ha-1 of dicamba (3,6-dichloro-o-anisic acid), and 1,000 g a.e. ha-1 of glyphosate, in addition to an additional treatment considered as a control. From the results obtained, it was not possible to observe the death of the plants submitted to simulated drift, regardless of the subdoses used and the stage of plant development. The simulated drift of dicamba alone and mixed with glyphosate in plants at the productive stage did not cause a reduction in productivity and yield. However, alterations in the classification of sieves were observed, with greater damage linked to the increase in subdoses used. Regarding the visual effects observed in seedlings, the action of dicamba herbicide presented similar symptoms to other agricultural crops. The main visual phytotoxic effects observed were foliar epinasty in the initial stage in the first pair of leaves with curvature towards the orthotropic branch of the plant, followed by alteration in the distance of internodes, with drastic deformation of the shape of young leaves in length and width, and necrosis at the leaf margins.