Investigation of spatial and temporal variability of saturated soil hydraulic conductivity at the field-scale.

摘要

Though soil hydrologists agree that field saturated conductivity (K_s) is a key parameter in modelling the dynamics of water flow and solute transport in soils, they also recognize that its variability in space and time is far from being completely understood. In order to highlight the variability of K_s at the plot scale we performed 10 measurement campaigns in three parcels within a 10 ha maize field during two subsequent crop seasons and in the fallow periods following them, in uniform conditions of crop, agricultural practices and, to a large extent, of pedological characteristics. This paper reports the outcomes of the measurements, conducted with the Guelph permeameter (GP) and with the tension infiltrometer (TI), along with detailed information on the data and a thorough description of the experimental field and of the measurement techniques. Based on a careful statistical analysis of the dataset and an extensive discussion of the results, the following conclusions were reached. GP K_s show changes in time and space, both between and within the parcels, with a different temporal behaviour for the different parcels, and no evident seasonal cycle. Mean and standard deviation of the transformed GP data samples are shown to be linearly related. This allowed the definition of a model of K_s statistical distribution that elucidates the distinct contributions of soil matrix and macropores, and provides a validation of the Morales et al. (2010. J. Hydrol. 393, 29) concept of biologically-driven macropore dynamics. TI estimations of K_s vary in space in agreement with the soil texture and show a stable seasonal pattern. However, in presence of macropores, they are not representative of the actual values of the saturated conductivity. On the other hand, TI K_s could provide an estimate of the conductivity of the soil matrix. The comparison with the soil matrix conductivity values deriving from the proposed model of K_s statistical distribution seem to support this possibility. These results, that shall be corroborated by further experiments, support the importance of thoroughly investigating the interactions between soil biota, vegetation and the soil hydraulic properties.