Evaporation experiments for the determination of hydraulic properties of peat and other organic soils: An evaluation of methods based on a large dataset

Soil hydraulic properties are frequently obtained from evaporation experiments. Different methods exist to infer soil hydraulic properties from these experiments. Commonly proposed is the 'direct method' (or 'Simplified Evaporation Method') by which soil hydraulic properties are calculated analytically. An alternative is given by inverse parameter optimization (Inverse method'). Although soil hydraulic properties are frequently estimated by the 'direct method', only very few studies have focused on the question how accurate derived parameters can reproduce laboratory measurements. This can be achieved by modeling the water flow with a processed based numerical forward model, e.g. with HYDRUS-1D. Here, we applied the 'direct method' and 'inverse method' to a large dataset of evaporation experiments on 431 organic soil samples. The derived soil hydraulic parameters were used in HYDRUS-1D simulations and their performance in reproducing measured states and fluxes was compared. As an additional analysis, we tested how water contents at the permanent wilting point can aid stabilizing parameter estimation by adding information on water retention in the dry range. For all methods, soil hydraulic properties were determined with the soil hydraulic functions of van Genuchten-Mualem and Peters-Dumer-Iden. The results show that parameters derived with the 'direct method' do often not well reproduce measured pressure heads over the complete pressure head range of the evaporation experiments when they were used for HYDRUS-1D simulations (mean objective function value 0.05831). Parameters derived by the 'inverse method' provided a better performance in the HYDRUS-1D simulations if the full pressure head range of the evaporation experiments was considered (mean objective function value 0.00099), but a weaker performance when focusing on wet conditions (pressure heads > - 100 cm). Constraining the 'inverse method' by additional soil moisture measurements at permanent wilting point improved the prediction of the soil moisture at dry conditions. For the full pressure head range, the hydraulic functions of Peters-Durner-Iden performed better than the ones of van Genuchten-Mualem.