Improving estimation of pore size distribution to predict the soil water retention curve from its particle size distribution

Traditional models employed to predict the soil water retention curve (SWRC) from the particle size distribution (PSD) always underestimate the water contents when the suction head is relative high. Using the measured physical parameters of 50 soil samples from the UNSODA unsaturated soil hydraulic property database, it is proven that this error originates from the poor conversion of the PSD to the pore size distribution (PoSD). The present study proposes a new approach to obtain a PoSD from a PSD, then to estimate the SWRCs. The residual pore fraction (the pore fraction with respect to clay content) was calculated using a power function, while the pore fractions of other sizes could readily be obtained using the corresponding particle mass fractions. The model was tested on 61 soils, and the results illustrated that the predicted PoSD from PSD using improved method exhibit good agreements with the measured PoSD, and the predicted SWRCs, which reduced the prediction bias at the relative high suction heads, reasonably approximated the measured SWRCs. Meanwhile, we evaluated the model performance through comparisons with the improved model proposed by Meskini-Vishkaee et al. (the scaled MV-VG model) and a traditional method, the results showed that the improved method provided better predictions of the SWRCs.