Study objectives are to (i) investigate the effects of a water-repellent top layer on water flow and solute transport and (ii) examine whether unstable wetting front theories can be used to predict the occurrence of preferential flow in water-repellent field soils. The study took place on two adjacent plots with grasscover near Ouddorp (The Netherlands)-one plot with a water-repellent top layer, the other plot with a wettable top layer. The soil at the experimental site is a sand soil of marine origin-mesic Typic Psammaquent. We conducted three different tests to obtain experimental evidence for our hypothesis that flow through a water-repellent soil would be unstable. An I- coloring technique showed an unstable wetting front with preferential flow paths in the water-repellent soil; flow through the wettable soil resulted in a homogeneous, stable wetting front. The variability of the soil water contents in the water-repellent soil was larger than in the wettable soil, indicating that flow through the water-repellent soil was less stable. A Br- tracer experiment, under conditions of natural precipitation, revealed that solutes in the water-repellent soil travelled faster to the groundwater than in the wettable soil. After 5 wk with 120 mm precipitation, the Br- amounts in the groundwater under the water-repellent top layer were 6 to 13 times higher than those under the wettable top layer. These experimental results demonstrate that unstable wetting front theories can be used to predict the occurrence of preferential flow in water-repellent field soils.
