Modeling soil water retention curves by pore fractal dimension based on microstructure image

Haiquan Sun
DOI: https://doi.org/10.1007/s12145-024-01367-7
2024-06-16
Earth Science Informatics
Abstract:The soil water retention curve is one of the most important characteristics of unsaturated soils, and it has a great relationship with soil permeability and mechanical characteristics. The laboratory measurement of the soil retention curve is quite a time consuming and expensive. In this article, four models, namely the Brooks-Corey (Hydrology Papers 1964) model, the van Genuchten (Soil Sci Soc Am J 44:1980) model, the Tyler-Wheatcraft (Soil Sci Soc Am J 53:1990) model, and the Rieu-Sposito (Soil Sci Soc Am J 55:1991) model, were introduced to predict the water retention curve based on the pore fractal dimension calculated from soil microstructure images with different magnifications. The four model predicted results were compared and discussed based on the root mean square error (RMSE) value. Results showed that the pore fractal dimension decreased with increasing magnification. The van Genuchten (Soil Sci Soc Am J 44:1980) model gives a better and more consistent prediction compared with other models. It proved that the van Genuchten (Soil Sci Soc Am J 44:1980) model is efficient for soil water retention curve prediction based on soil microstructure images analysis.
geosciences, multidisciplinary,computer science, interdisciplinary applications
What problem does this paper attempt to address?