Transport from diffuse sources of contamination and its application to a coupled unsaturated - saturated system

A simple theory to predict groundwater quality upon contamination from diffuse sources was developed. It appeared that an analogy exists between the predominant transport phenomena and the reaction of a reservoir, in which perfect mixing takes place. Such an analogy enables a simple incorporation of physico-chemical processes (decomposition, adsorption), as was shown by an illustrative response of the quality of groundwater to an input of a radio-active decaying solute (and its decay components) in its recharge area.Before solute migration in the saturated zone occurs, transport takes place in the unsaturated zone. The above mentioned analogy allows the derivation of analytical expressions for the response of a coupled unsaturatedsaturated transport system. The unsaturated soil is then represented by a series of "perfectly mixed" reservoirs. Each reservoir can be defined in terms of its thickness, water content, inand outflow rate, decomposition rate or retardation factor. Moreover, the process of bypass flow can be accounted for in a simple way.To study the practical application of the presented theory, a field experiment was carried out. An amount of tracer was applied to the catchment area of a drain, followed by soil sampling and monitoring of tracer concentrations in the drainage water. From both soil and drainage water sampling it appeared, that transport velocities in the subsoil could impossibly correspond with the observed water content profiles. Therefore it was concluded that preferential flow paths in the subsoil caused this accelerated breakthrough, and transport was confined to these zones only.To visualize these zones of preferential flow, we used a coloring technique that was developed during this study. This method was applied to ten plots in the same experimental field, where the tracer had been applied before. Preferential flow paths were shown to occur at some plots, but the variation of penetration depth of the solute front could be adequately described by the convection-dispersion equation, using dispersivities of the order of centimetres. The accelerated breakthrough that was measured in our former experiment, could not be explained.At the same experimental field, a study of the spatial variability of soil physical parameters had been completed recently. From the measured variability, expressed in the distribution of the scaling parameter, the solute breakthrough curve was predicted, and compared with our measured curve. The variation of travel times, based on the soil variability, was estimated too large. It is concluded that this large variation of travel times might have been present under different experimental conditions, such as ponded infiltration.