Evaporation from soil surface is the main driving force of salinization of irrigated soils in arid and semi-arid land. To optimize irrigation scheduling under saline conditions, it is essential to accurately predict solute transport in soils and an evaporation rate. We conducted laboratory column experiments under constant meteorological conditions except for radiation, which was automatically regulated such that temperature of the soil remained constant at that of air. Concentration of initial and inflow water from the bottom was set at 3,000 ppm and height of the column was 5.2cm. The evaporation experiments were performed with three combinations of soil and solute: a sand & NaCl, a loamy sand & NaCl, a loamy sand & KCl. While the soil surface was kept wet by keeping suction at the bottom low, the evaporation rate was considerably decreased with time. This decrease cannot be explained by osmotic potential alone. We thus added the resistance to water vapor diffusion due to salty crust into bulk transfer equation, and evaluated the dependence of the resistance term, salty crust resistance, on mass of accumulated salt. Using the salty crust resistance, we calculated solute movement and evaporation rate. The Richards' equation and convection dispersion equation (CDE) were solved with the finite difference method. Water vapor movement and the effect of osmotic potential were also incorporated. Independently obtained hydraulic and solute transport parameters were used in the analysis. Results showed that the CDE tends to overestimate backward diffusion near the evaporating soil surface, which resulted in significant delay of both accumulation to soil surface and decrease in evaporation rate. Since the CDE uses an analogy of the Fickian law to describe the mechanical dispersion, the dispersion term would overestimate downward movement of solute against upward convective transport.