Predicting water retention curves of fine texture soils from particle size distribution

Indirect methods for estimating soil hydraulic properties from particle size distribution have been developed due to the difficulty in accurately determining soil hydraulic properties, and the fact that particle size distribution is one piece of basic soil physical information normally available. The similarity of the functions describing the cumulative distribution of particle size and pore size in the soil has been the basis for relating particle size distribution and the water retention function in the soil. Empirical and semi-physical models have been proposed, but these are based on strong assumptions that are not always valid. For example, soil particles are normally assumed to be spherical, with constant density regardless of their size; and the soil pore space has been described by an assembly of capillary tubes, or the pore space in the soil matrix is assumed to be arranged in a similar way regardless of particle size. However, in a natural soil the geometry of the pores may vary with the size of the particles, leading to a variable relation between particle radius and pore radius.

 

The current work is based on the hypothesis that the geometry of the pore size and the void ratio depends on the size of the soil particles, and that a physically based model can be generalised to predict the water retention curve from particle size distribution. The rearrangement of the soil particles is considered by introducing a mixing function that modulates the cumulative particle size distribution, while the total porosity is constrained by the saturated water content.

 

The model performance is evaluated by comparing the soil water retention curve derived from laboratory measurements with a mean Nash–Sutcliffe model efficiency a value of 0.92 and a standard deviation of 0.08. The model is valid for all soil types, not just those with a marginal clay fraction.