Effect of laboratory procedures on the integrity of a sandy-loam soil sample for bi-directional measurement of saturated hydraulic conductivity

Practical and validated methods to measure anisotropy of saturated, Ks, or field-saturated, Kfs, soil hydraulic conductivity on a single soil sample are still lacking. The objective of this investigation was to test factors affecting anisotropy measured in a sandy-loam soil by the constant-head laboratory permeameter and the modified cube method. The ratio, RK, between the conductivity results obtained with a long (6 h) and a short (0.5 h) duration run varied from a minimum of 0.29 to a maximum of 0.88 with the considered experimental procedure, differing by both the employed constant-head device (Mariotte bottle, siphon) and the initial soil water content (unsaturated, saturated). Maximum time stability was detected with a siphon and an initially saturated soil sample, suggesting that soil alteration at the surface may influence the measured conductivity, and more reliable anisotropy measurements have to be expected for Ks than Kfs. A short-duration run carried out with a siphon and a calcium solution was also suggested for laboratory determination of Ks anisotropy since more representative results were obtained for a single cube of soil (i.e., horizontal, Kh, to vertical, Kv, conductivity ratio varying from 1.78 to 1.82 with the measurement sequence) as compared with the ones associated with using tap water (ratio varying from 1.24 to 2.09). The Kh/Kv ratio varied between 1.80 and 2.81 on two dates (April and July 2009) differing by the soil water content at the time of sampling (close to field capacity; lower than the permanent wilting point, respectively), and a significant temporal variability was only detected for Kv. Therefore, temporal variability of Ks anisotropy cannot be predicted by measuring one-dimensional Ks.