Use of a lognormal distribution model for estimating soil water retention curves from particle-size distribution data

Particle-size distribution (PSD) of soils has been widely used to estimate water retention curves (WRCs). Lognormal distribution model, which estimates the WRC directly from the PSD data by applying a lognormal distribution law to both PSD and void-size distribution (VSD), has been recently developed and successfully applied to sandy soils. The objective of this study was to investigate in detail the applicability of the lognormal distribution model for a broader range of soil textures. Performance of the model was evaluated on five aggregated USDA texture classes (coarse-, moderately coarse-, medium-, moderately fine-, and fine-texture soils) using 229 ‘undisturbed’ soils (159 soils for model calibration and 70 soils for validation). The model performed worst for moderately fine-textured soils, since the PSDs and VSDs of soils included in this texture did not strongly follow the lognormal distribution. In general, performance of the model decreased with increase in fine particle fraction, possibly because of the surface sorption behavior of clay particles.