Two-dimensional modelling of preferential water flow and pesticide transport from a tile-drained field

Abstract

Summary Preferential flow through soil macropores in tile drained soils can significantly increase the risk of pollution of surface water bodies by agricultural chemicals such as pesticides. While many field studies have shown the importance of preferential flow in tile-drained fields, few have included detailed numerical modelling of the processes involved. The objective of this study was to compare four conceptually different preferential flow and/or transport approaches for their ability to simulate drainage and pesticide leaching to tile drains. The different approaches included an equilibrium approach using modified hydraulic properties near saturation, and three non-equilibrium approaches: a mobile–immobile solute transport model, a dual-porosity approach, and a dual-permeability formulation. They were implemented into the HYDRUS-2D software package. The model predictions were compared against measurements of drainage and pesticide concentrations made at an undulating, tile-drained field in southern Sweden (Nasbygard) during a period of 6 weeks following spray application of the herbicide MCPA. The dual-permeability approach most accurately simulated preferential drainage flow, even though this approach somewhat overestimated the drainage rates. The equilibrium and mobile–immobile approaches largely failed to capture the preferential flow process. The dual-porosity approach predicted much more distinct and higher drainage flow events as compared to the dual-permeability approach. Differences between measured and simulated tile drainage rates using the dual-permeability approach could be partly explained by water by-passing the tile drains and recharging the deeper aquifer. The dual-permeability and dual-porosity approaches closely captured the dynamics in measured pesticide concentrations. Both the equilibrium and mobile–immobile approaches completely failed to match measured MCPA leaching by underestimating the peak concentrations by more than two orders of magnitude. We conclude that two-dimensional models are suitable tools for studying pesticide leaching from undulating fields with large spatial variability in soil properties.