Search

The capillary effect in polydisperse porous systems of soils and sediments is discussed. This effect is quantified using a new fundamental model of capillary rise depending on the parameters of dispersion of soil particles, soil bulk density, density of soil solid phase, wettability, and the amount of tightly bound water. In contrast to a well-known Jurin's law, the model takes into account the particle-size distributions and the change in pore size associated with soil density. The model adequately predicts the capillary rise in both artificial monodisperse porous systems and real soils with the particle size ranging from 0.006 to 1 mm. The effects of natural and synthetic hydrophobizers on the kinetics and maximum height of capillary rise are examined. The laboratory and field data on the efficiency of imperfect (with hydraulic connection) and perfect (with its rupture) capillary barriers are compared to the Hydrus-1D predictive modeling data. The results are used to evaluate the capillarity in soils and to design soil constructions with capillary barriers. [ABSTRACT FROM AUTHOR]

The specific surface area (SSA) of a soil is crucial for the interface of ions and water molecules with the soil particles. Therefore, many physical and chemical properties of a soil are determined using its total SSA. Measurement of SSA is time-consuming and laborious, and its estimation using pedotransfer functions is therefore preferred. The objectives of this study were to: (1) analyze the pore–solid interface fractal dimension (D) in soils with different textures from thesouthern part of Iran using estimated improved particle-size distribution (PSD) from the soil primary particles, i.e. clay, silt and sand; (2) develop a multivariate pedotransfer function to estimate D based on PSD; and (3) develop a multivariate pedotransfer function to link the values of D to the SSA. As a result, two pedotransfer functions are presented for estimation of D and SSA. To estimate SSA, the value of D is first obtained using the presented pedotransfer function and these estimated D values are then used in another pedotransfer function to estimate SSA. The pedotransfer functions were validated and it is concluded that they are able to predict the values of D and SSA accurately. [ABSTRACT FROM PUBLISHER]