Simulating surface and subsurface initiation of macropore flow

Initiation of macropore flow either from the soil surface or from a saturated soil layer at depth is a first order control on water flow in macropores and water transfer from macropores into the surrounding soil matrix. Nevertheless, these initiation processes have not been well documented. We surveyed surface topography at four field sites with permanent grass vegetation with grid spacing of 10 cm and applied Kriging to derive the spatial correlation structure. We then simulated the water flux into macropores based on different combinations of surveyed surface micro-topographies, spatial earthworm burrow distributions, and the soil properties, to examine more fully the role of macropore drainage area (MDA) on macropore flow initiation. The spatial distributions of the earthworm burrows were derived from horizontal soil sections extracted from each study profile. The MDA was calculated for different sets of surface topography and macropore density using a flow accumulation algorithm. The resulting MDA of each macropore was used to calculate the total relative NMA, which is equal to the proportion of overland flow draining into macropores, and the MDA probability distribution. The results showed that the macropore density primarily controlled the total MDA and that surface micro-topography strongly influenced the probability distribution of the MDA. Only a few macropores contributed significantly to the total macropore flow whereas the majority of macropores received little water; a phenomenon especially pronounced for a rough surface topography and for a low soil surface gradient. The simulated probability distribution of subsurface initiation was very different from the distribution derived for surface initiation; more symmetrical, less variable and slightly influenced by the roughness and the gradient of the interface between the saturated and the low permeable soil layer. We conclude that the different amount of water supplied to each macropore further alters the percolation depth and transport of solutes in macroporous soils and should be considered for modelling infiltration in macroporous soils. (C) 2003 Elsevier Science B.V. All rights reserved.