Prediction of water retention curve of a layer packed with soil particles from particle size distribution
Faculty of Computing, Health and Science
School of Natural Sciences
This study developed a method to predict the water retention curve of a layer packed with soil particles of various sizes from the particle size distribution and the voidage of the packed layer. The results indicated that the predicted water retention curve coincided well with the measured one in the case of a packed layer of spherical particles having a narrow size distribution. Also, the results showed that the developed method could be applied to a packed layer of nonspherical particles like Toyoura sand, which also had a relatively narrow particle size distribution. In the case of soil particles with a wide size distribution, the predicted water retention curve showed much lower water content than the measured one at lower water potential. This indicated that the contribution of larger particles to a change in water content tended to be overestimated in the developed method. The method was therefore modified to reduce the contribution of larger particles, and as a result, the water retention curve predicted by the modified method agreed well with the measured one.