摘要：The relationship between capillary pressure and saturation has a remarkable value in investigating interactions of two immiscible fluids in porous media. Pore network models, as predictors of fluids movement in porous media, can play undeniable role in determining the mentioned relationship. In the current study, the results of numerical pore network modelling, which represents pore structure of a real porous medium with networks of pore bodies connected with pore throats, are used for computing the macroscopic relationship between capillary pressure and saturation. A notable volume of soil is influenced by water source and according to the results of previous studies, it seems practically impossible to simulate water movement in macro scale dimensions by use of pore scale models. Meanwhile a suitable solution was found in this study, by considering a thin vertical soil column under the water source which was divided to smaller volumes as cells, by horizontal crossings. Each cell was considered as a pore network unit, so the soil column was consisted of series of pore network units which were vertically jointed to each other. The moisture distribution and also wetting front movement in the column were determined by application of pore network model, using the dynamic update saturation solution method. The solution was conducted between each pair of consequent cells in an alternative manner within a time step since arrival time of the water from an upper cell to the lower one. Moreover, for evaluation of the model ability, soil moisture profiles in a sandy soil of an experimental tank under the water source were studied. Comparison of the simulation and observation data confirmed the high ability of the column pore network model in prediction of the moisture distribution and wetting front movement in a soil column.
