Scaling Water Retention Curves of Dissimilar Soils

Scaling methods have been developed to describe the spatial variability of soils. Among them, physically based methods are more desirable in which the scaling factors can be estimated from the soil physical properties. Assuming that the pore size distribution of soils is lognormally distributed, Kosugi and Hopmans developed a physically based method to scale the water retention curve. However, similar to the previous methods, application of this method is limited to the similar soils. To alleviate this limitation, in this paper, a physically based method is proposed for dissimilar soils. Using this method, data of a wide textural range of soils can be scaled without the similarity condition and can be represented by a unique exponential reference curve. This method was validated using 487 sets of soil retention curves taken from UNSODA database including all the textural classes from sand to clay. The results showed that the proposed method had a better performance in scaling the retention curves than that of Kosugi and Hopmans. The criteria defined here for scaling error was obtained equal to 0.074 and 0.105 for the proposed and Kosugi-Hopmans method, respectively. In addition, it was shown that, in contrast to the previous methods, the scaling error of the proposed method does not depend on the soil texture and all the soils have an equal chance for being scaled.