Effects of simulated rainfall intensity on water flow through soil in two tillage systems

Diffuse losses of phosphorus from agricultural land to freshwater reservoirs have gained increasing interest in later years. Resent research suggests that phosphorus is mainly transported from the fields through macropores, and that soil under reduced forms of tillage can develop a higher macroporosity and thus an increased risk of phosphorus leaching compared to conventionally tilled soil, because of the more rapid flows of water through the soil. In this study field measurements of flow responses to variations in rainfall and laboratory measurements of soil dry bulk density, soil porosity and soil water retention characteristics were compared for a heavy clay soil subject to reduced and conventional tillage respectively. In the field, rainfall with intensities of 10 or 33 mm/h were simulated and percolating water was collected from a cavity dug out at 40-45 cm depth and the collected volume was measured at given times. A simple dual-permeability model was also made and its ability to describe flow responses to changes in rainfall intensity was evaluated through comparison with the field measurements. Flow responses to variations in rainfall intensity were rather similar between the two tillage treatments. Stabilized outflow rates were also very similar between the treatments; between 4 and 8.5 mm/h under the lower rainfall intensity and between 22 and 28 mm/h under the higher. Neither of the parameters measured at the lab showed any clear differences between the two tillage treatments. The model was able to describe the timing of changes in outflow in response to changes in rainfall intensity, but the simulated outflow rate was higher than what was measured and the model also described different final soil water contents after rainfall from those that were measured. Overall, the measurements did not show any effect of tillage treatment on soil hydraulic properties on the profile scale; flow showed dependence mainly on rainfall intensity and soil moisture conditions. The model would need to account for water flow from macropores to matrix to better be able to describe the total outflow from the profile.