Determination of soil water hydraulic parameters from infiltration data

Abstract

PurposeTraditional laboratory measurements of soil water diffusivity (D) and soil water retention curve (SWRC) are always time-consuming and labor-intensive. Therefore, this paper aims to present a simple and robust test method for determining D and SWRC without reducing accuracy.Design/methodology/approachIn this study, a D model of unsaturated soil was established based on Gardner–Russo model and then a combination of Gardner–Russo model with one-dimensional horizontal absorption method to obtain n and a parameters of Gardner–Russo model. One-dimensional horizontal absorption experiments on loam, silt loam and sandy clay loam were conducted to obtain the relationships between measured infiltration rate and cumulative infiltration with wetting front distance. Based on the obtained relationships, the measured infiltration data from the one-dimensional horizontal absorption tests were used to calculate n and a parameters and further constructing D and SWRC.FindingsBoth the calculated D and SWRC inversed from the infiltration data were in good agreement with the measured ones that obtained from the traditional horizontal absorption method and the centrifuge method, respectively. Error analysis indicated that only the infiltration data are enough to reliably synchronously determine D and SWRC.Originality/valueA simple and robust method is proposed for synchronous determination of soil water diffusivity and water retention curve.

Soil Infiltration Rate Impact on Water Quality Modeled Using Random Forest Regression

“…The network is changed in accordance with the output-to-target ratio until the total of the squared discrepancies between the target and output values equals zero. Typically, the network is trained using numerous such target input/output pairs [20]. Following the presentation of each individual input vector, more training is performed to alter the weights and biases of the network as appropriate.…”