Simulation of chloride transport in an aggregated soil using three conceptual models

Abstract

Breakthrough curves (BTCs) of chloride displaced through columns of loessial soil aggregates of different sizes were measured under saturated steady flow conditions. The data were simulated using three conceptual models. Model I (CDE) assumed that all soil water was mobile and physical equilibrium existed in the system. Model II (two-region model) partitioned the soil water into mobile and immobile regions, and convective diffusive solute transport was limited to the mobile water region. Model III (two-flow region model) also divided the soil water into two regions based on their flow velocities, but both of the regions had a non-zero flow rate. Transfer of the chloride solute between the two soil water regions was assumed to occur at a rate proportional to the difference in solute concentration. The two unknown parameters in model I, three in model II, and four in model III were estimated by fitting the experimental data. The three models could well describe all the BTCs measured for columns packed with all the aggregate sizes at the low pore water velocity (0.68 cm/h); however, the values of the fitted parameters varied greatly. The Peclet numbers derived from both the two-region (model II) and two-flow region (model III) models behaved similarly and increased with increases in aggregate size. But the Peclet numbers derived from the convection dispersion equation (model I) were about two orders of magnitude greater than those derived from the other two models. The mobile water fraction obtained for the two-flow region model decreased with increases of aggregate size. The mass transfer coefficient decreased with an increase in pore water velocity due to the shorter residence time of the chloride solute in the soil columns.