The influence of the measurement method on the values of the mean and standard deviation of saturated hydraulic conductivities, Ks, has not been extensively studied. This study examines the dependence of these statistical characteristics of Ks on the sampled soil volume. Statistical characteristics of Ks for a sandy loam soil determined by field and laboratory techniques differed greatly. However, if anisotropy in hydraulic conductivity induced by soil structure and macropores is considered, the mean of field vertical Ks (Kvs) approaches that of laboratory Ks. Furthermore, geostatistical variograms of log(Ks) and the regression analysis (field/laboratory) demonstrate the spatial independence of Ks. Thus, when generating Ks values for larger soil cores, standard deviation approaches field Kvs. When the investigated areas are the same for both methods, mean and scattering of field-and laboratory-measured values become closer. These results show that the method as well as the measurement area greatly affect the representativeness of measured Ks values and their scattering.
Read full abstract