Dual-permeability modeling of preferential bromide leaching from a tile-drained glacial till agricultural field

摘要

A tile-drained agricultural field can be regarded as a 'field-scale' lysimeter that maybe used to study soil water and chemical transport under relatively natural conditions. Tile discharge and effluent bromide concentrations measured in a previous field tracer experiment for a structured clayey loam at Bokhorst, Northern Germany, indicated strong preferential flow. Simulation using single domain HYDRUS numerical flow and transport model could nevertheless explain water outflow, however, completely failed to describe tile-drain leaching patterns of the conservative tracer. The objective of this paper was to analyze whether the nonequilibrium-type dual-permeability model concept could better capture soil structure related principle mechanisms of preferential leaching in the unsaturated soil at that study site. The dual-penneability model (DUAL) describes for soil matrix and fracture pore systems Darcian flow with coupled Richards' equations and convective-dispersive (CD) solute transport with coupled CD equations. The hydraulic parameters of the dual-permeability model were obtained from standard soil hydraulic measurements by adopting a bimodal fitting procedure, whereas transport parameters were inferred front soil column tracer experiments and geometrical transfer term parameters were derived using qualitative soil structure descriptions. The hydraulic conductivity K,, in the inter-domain water transfer term and the effective diffusion coefficient D. in the solute mass transfer term were calibrated by comparing simulated with measured tile flow and effluent concentrations. The DUAL approach described water flow similarly well as the single-domain model, Bromide concentrations in the tile effluent could be approximated with DUAL when decreasing the K, and D, values by three orders of magnitude compared with the values of the soil matrix domain. The dual-penneability approach seems to reflect nonequilibrium transport mechanisms at this structured soil since it not only predicts concentration peaks in initial leaching phases during early time events of the experiment but also the observed reverse reaction of decreased effluent concentrations in initial phases of leaching events during later times of the experiment. The latter occurs when bromide-free infiltrating water is diluting bromide concentrations in the fracture domain such that less concentrated water is preferentially draining from tiles in initial phases of infiltration. Bromide effluent concentrations gradually increase in later phases since vertical flow velocities decrease while 'diffusive' solute transfer from the soil matrix continues. Mass transfer restrictions (e.g. due to soil aggregate coatings) may effectively be controlling preferential leaching at this site. (C) 2004 Elsevier B.V. All rights reserved. [References: 73]