In situ estimation of soil hydraulic functions using a multiple step soil-water extraction technique

Estimation of the retention and unsaturated hydraulic conductivity functions is essential to effectively provide input for water flow and transport simulation and prediction. A parameter optimization procedure is shown as a promising tool to estimate inversely these hydraulic function parameters from transient soil matric potential and cumulative soil solution extraction measurements. Sensitivity analyses from synthetic data generated from forward numerical model simulations showed that optimum tensiometer locations will depend on soil type. Experiments were carried out in both a laboratory column (Columbia sandy loam) and in the field (Yolo silt loam). In both cases a series of vacuum extraction pressures was applied to a ceramic soil solution sampler, and cumulative soil solution extraction volume and matric potentials at various positions near the extraction device were monitored as the soil solution was extracted. In the laboratory a zero-flux boundary condition was maintained at the bottom of the column, whereas matric potential measurements were used in the field to define the lower boundary. In both the field and laboratory experiments, flow at the upper boundary was zero. Cumulative extraction volume and matric potential data were included in the objective function to be minimized to estimate the hydraulic function parameters. We determined that the optimized solution was sensitive to the contact between the ceramic ring and the surrounding soil. By also optimizing the hydraulic resistance of the ceramic ring of the extraction device, optimization improved the fit between measured and optimized flow variables. Comparison of the optimized with the independently measured hydraulic functions indicated that the in situ estimation using a multistep extraction procedure can provide accurate soil hydraulic data.