A soil-brine retention model for wetting processes considering the hysteresis effects

Abstract

Wetting hydrological processes are relevant to a considerable number of natural hazards, including landslides, dam breaks, surface runoff after floods, etc. The notable difference between wetting and drying soil–water retention curves (SWRC) has been attributed to the hysteresis phenomenon. Although recent studies confirm enhanced impurity levels of permeating fluid, most modeling approaches simply consider water retention along the drying path and neglect the impurity of pore water. Therefore, the main objective of this paper is to introduce a new water retention model along the wetting path by taking the osmotic potential into account, reflecting the influence of dissolved salts. The new model, referred to as the soil-brine retention curve (SBRC) was developed based on the 1980 van Genuchten model (VG). The results reveal the robustness of the model in predicting the experimental wetting retention data for different soil types exposed to both saline and distilled water. Statistical analysis confirms the capability of the new model with an average R2 of 0.97 and RMSE of 0.047. The wetting branch can be estimated based on an existing drying curve, assuming predefined values of entrapped air and air expulsion value. However, a more reliable prediction can be made if two extra parameters of entrapped air and air expulsion are calibrated for the pure water wetting curve. Furthermore, for the case of saline water, the wetting SBRC can be estimated using one more parameter than those used for pure water.