Estimating unsaturated soil hydraulic properties in sloping lands by numerical inversion

摘要

Inverse solutions used an optimization algorithm to estimate the unknown parameters. Tension disk lnfiltrometer is a suitable instrument for determining the soil hydraulic properties such as hydraulic conductivity. In this study the tension disk infiltrometer data were used to estimate the unknown parameters of VanGenuchten (VG) soil water retention and Mualem-VanGenuchten (MVG) hydraulic conductivity models in different slope gradients including 0, 10, 20, 30 and 40 degree of slope. The unknown parameters, including 0, 0 s, (X, /, n and Ks have been estimated by numerical inversion method using Hydrus 3D code. Initial estimates of unknown parameters were obtained from Rosetta code (for estimating 0^ 0s, a, /, n and Ks) and Mualem report (for estimating I). Field experiments were carried out in a loamy soil with different slope gradients in Gonbad Research Station, Hamadan, Iran. For each slope gradient, infiltration experiments were conducted at tensions of 0, 6,9 and 15 cm, using a tension infiltrometer disk. Totally 60 infiltration experiments in 5 treatments (slope gradients), 4 tensions and 3 replications were carried out. Results indicated that with increasing slope gradient, 0s, Ks and n values were decreased, 6r values increased and (X and/ values had no regular trend. By increasing tension, the falling rate of decreased with increase in slope gradient. Unsaturated hydraulic conductivity also decreased with increase in slope gradient. The (9and lvalues were estimated by numerical inversion method less than that measured for all slope gradients. The hysteresis effect and dependence of unsaturated hydraulic parameters together caused that numerical inversion underestimated the 0 and K values less than measured data.