Glyphosate behavior in soil and miscible displacement of atrazine.

Author(s)

Fabio Prata

• Document type: Doctoral Thesis
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2002-09-06

Examining board members

Arquimedes Lavorenti; Pedro Jacob Christoffoleti; Rosa Toyoko Shiraishi Frighetto; Luiz Carlos Luchini; Maria Emilia Mattiazzo Prezotto

• Advisor: Arquimedes Lavorenti

Abstract

The aim of this research was to evaluate the glyphosate behavior in soils under different aspects, as well as the miscible displacement of atrazine, using mathematical modeling to determine sorption and transport parameters. The experiments about glyphosate were carried out at Laboratory of Ecotoxicology of the Centro de Energia Nuclear na Agricultura (CENA/USP), Piracicaba, SP (Brazil), and the studies about atrazine at Forchungszentrum Jülich, Jülich, Germany. In the first chapter we evaluated the effects of soil organic matter on the sorption and desorption of glyphosate in three soils with different mineralogical attributes. This was an isotherm batch experiment in soil samples with or without organic matter. In the second chapter we evaluated the effects of the increasing rates of phosphorus on the sorption and desorption of glyphosate. In the third chapter we studied the glyphosate behavior as an especial situation: An Brazilian Oxisol collected from both a no-till and a conventional management soil systems. Both agricultural systems had been in production for 23 years. Glyphosate mineralization, its bound-residue forms, sorption and desorption batch kinetics, sorption/desorption batch isotherms experiments, and glyphosate phythoavailability (to Panicum maximum) were determined. In the last chapter, the miscible displacement and the sorption/desorption of atrazine. In order to fit the atrazine breakthrough curves, to evaluate the contribution of sorption on the atrazine leaching (equilibrium vs nonequilibrium sorption; reversible vs irreversible sorption), we employed a tracer study (Br - ) and used mathematical modeling. We fitted the Br - displacement with an equilibrium convective dispersive transport model, and atrazine displacement with two-site chemical nonequilibrium dispersive convective transport model. The atrazine was employed in this study, instead glyphosate, because the second one does not present leaching potential. We concluded the glyphosate sorption is instantaneous, extremely high and presents relationship, mainly, with the mineral soil fraction. Soil organic matter only plays a secondary role for oxidic soils. Glyphosate competes with inorganic phosphates for specific sorption sites, but this competition becomes important when phosphorus is at rates higher than the normally seen in real field conditions. The glyphosate can not be extracted from the soil, under normal conditions of agricultural soils, remaining as bound-residues. The no-till system may contribute to the enhancement of glyphosate mineralization in the soil. However, its half-life is low and is related with the bound-residue formation. Glyphosate did not present phytotoxicity in the specific studied. About atrazine displacement, we concluded the atrazine presented a potential leaching. This potential was dependent of equilibrium and nonequilibrium sorption sites. However, irreversible sorption was more important than "reversible" sorption in equilibrium and nonequilibrium sites. Both equilibrium and nonequilibrium models successfully fitted the Br - and atrazine breakthrough curves, respectively, for both management systems. However, the two-site nonequilibrium model overestimated the irreversible sorption (bound-residue formation) for resident concentration of atrazine within soil column. The two-site nonequilibrium transport model predicted partition coefficients for sorption very similar to these measured in the sorption batch experiment. This was not true for desorption data. (AU)