Comparing mini-disk infiltrometer, BEST method and soil core estimates of hydraulic conductivity of a sandy-loam soil

Saturated, K_s, and near-saturated, K, soil hydraulic conductivity control many hydrological processes but they are difficult to measure. Comparing methods to determine K_s and K is a means to establish how and why these soil hydrodynamic properties vary with the applied method. A comparison was established between the K_s and K values of a sandy-loam soil obtained, in the field, with the BEST (Beerkan Estimation of Soil Transfer parameters) method of soil hydraulic characterization and an unconfined MDI (mini-disk infiltrometer) experiment and, in the laboratory, with a confined MDI experiment and the CHP (constant-head permeameter) method. Using for the BEST calculations the soil porosity instead of the saturated soil water content yielded 1.4–1.1 times higher estimates of K_s and K, depending on the pressure head, and differences decreased in more unsaturated soil conditions. The confined MDI experiment yielded 22 % - 77 % higher K values than the unconfined MDI experiment, depending on the established pressure head, h₀, and differences were not significant for h₀ = −1 cm. In the close to saturation region, the soil hydraulic conductivity function predicted with BEST did not generally agree well with the K_s and K values obtained in the laboratory by a direct application of the Darcy’s law. In particular, BEST yielded a 5.6 times smaller K_s value than the CHP method and up to an 8.1 times higher K value than the MDI. Overall, i) the two application methods of the MDI yielded relatively similar results, especially close to saturation, and ii) there was not a satisfactory agreement between the field (BEST) and the laboratory (MDI plus CHP) determination of soil hydraulic conductivity close to saturation, unless a comparison was made with the same soil water content. The detected differences were probably attributable to soil spatial variability, overestimation of K_s in the laboratory due to preferential flow phenomena, underestimation of K_s in the field due to air entrapment in the soil and infiltration surface disturbance, inability of BEST to describe the actual soil hydraulic conductivity function at the sampled field site. Testing BEST predictions of K_s and K in other soils appears advisable and combining the MDI and CHP methods appears a rather simple means to make these checks. These additional investigations could improve interpretation of the differences between methods, which is an important step for properly selecting a method yielding K_s and K data appropriate for an intended use.