Metribuzin transport in undisturbed soil cores under controlled water potential conditions: experiments and modelling to evaluate the risk of leaching in a sandy loam soil profile

BACKGROUND: Mobility of pesticides in soils is often evaluated and characterised in the surface soil layers rather than at different depths where soil characteristics such as soil organic matter, microbial biomass or clay contents can strongly change pesticide behaviour. The objective of this work was to characterise the reactivity of the herbicide metribuzin in three main soil horizons found in the 0-80 cm profile of an alluvial soil of southern Norway under dynamic transport conditions. RESULTS: A laboratory infiltrometer was used to perform percolation experiments in soil cores sampled in the three horizons Ap, Bw and Bw/C, at a fixed matric potential of -10 cm, thus preventing pores of equivalent radii higher than 0.015 cm from contributing to water flow. The physical equilibrium transport model correctly described the transport of water tracer (bromide). The distribution coefficient K(d) values were estimated to be 0.29, 0.17 ± 0.02 and 0.15 ± 0.00 L kg(-1) for horizons Ap, Bw and Bw/C respectively, in close agreement with batch sorption data. Degradation was found only for the surface horizon with a short half-life of about 5 days, in disagreement with longer half-lives found in batch and field degradation data. CONCLUSION: For all horizons, a kinetic sorption model was needed for better description of metribuzin leaching. Chemical non-equilibrium was greatest in the Bw horizon and lowest in the Bw/C horizon. Overall, metribuzin exhibited a greater mobility in the deeper horizons. The risk of metribuzin transfer to groundwater in such alluvial soils should therefore be considered.