Comparison of the Instantaneous Profile Method and inverse modelling for the prediction of effective soil hydraulic properties

Abstract

Soil hydraulic conductivity K(θ) and soil water retention θ(h) have been determined from a drainage experiment. Two lysimeters, one filled with a sandy soil and the other with a loamy soil, were set up for a 1-dimensional transient flow experiment. The data were collected after flooding the lysimeters with water. Soil water contents were measured by time domain reflectrometry (TDR) and pressure heads were measured by tensiometers with mercury manometers. The experimental data determined by the instantaneous profile method (IPM) were compared with the results obtained by inverse modelling. The inverse modelling proved to be superior to the IPM methodology in effective prediction of hydraulic properties. The measurable properties water content and pressure head were optimised for the following datasets: water content (WC), pressure head (P-h), and a combination of WC and P-h. For both soils the optimisation of the dataset with both WC and P-h resulted in parameters that corresponded closely to the soil hydraulic data generated by the IPM method. The correspondence for the water retention data was better than for the hydraulic conductivity data. The datasets with WC only or P-h only did not contain enough information to accurately estimate the soil hydraulic properties. In most cases the results indicated that the sandy soil gave better agreement than the loamy soil. This was attributed to the faster drainage of the sandy than the loamy soil.