Laboratory and field soil hydraulic characterization by parameter estimation approach.

In this paper, an inverse method for the determination of unsaturated hydraulic properties from transient flow events is developed and applied in the laboratory and in the field to a number of soils. For the laboratory method, variables considered in the inversion procedure are pressure potential at various locations and soil sample weight as a function of time during an evaporation experiment. The determination of in-situ soil hydraulic properties involves measurements of water content and pressure potential during a drainage experiment in a small plot. Two different parametric models are used for describing soil hydraulic properties: van Genuchten's well-known model (the VG model), and a model that couples the water retention expression suggested by van Genuchten with an exponential relationship between hydraulic conductivity and water content (the EXP model). A non-linear least-squares optimization problem is formulated to estimate unknown parameters in the constitutive hydraulic models by minimizing the deviations between the numerical solution of the transient flow process and the real system response measured during the experiment. Information concerning the uncertainty in parameter estimates is provided as well. The accuracy of the proposed methods is shown by comparing estimated water retention and hydraulic conductivity curves with data points obtained via the instantaneous profile method. Additionally, stability of the inverse solutions is discussed with reference to the original methods and to the modified methods, which were developed with a view to reducing experimental efforts. Results from both practical and numerical examples show that the optimization algorithm is accurate and cost-effective, and suggest that these methods can be successfully applied for spatial variability studies which require a large number of observation points.