Coupled water and heat flow in a grass field with aggregated Andisol during soil-freezing periods

During soil-freezing periods, coupled water and heat flow is important for predicting frost depth and unsaturated water flow between frozen and unfrozen soil. We investigated water and heat flow in Andisol with aggregated soil structure at a grass field during soil-freezing periods. The water retention curve (WRC) had a stepwise shape, in which water content, θ, decreased drastically at air entry value, h = −0.3 m, and matric potential, h = −10 m. The profiles of θ and temperature, T, in an Andisol were measured using thermally-insulated tensiometers and thermo-time domain reflectometry (thermo-TDR) probes in the northeastern part of Japan. As the surface soil froze, soil water moved upward because of the matric potential gradients. Although unfrozen water content, θ, in water-saturated frozen soil may be described using the generalized Clausius–Clapeyron theory, the theory cannot express θ in unsaturated frozen soil. When soil started to freeze in unsaturated conditions at a field, the matric potential may be a significant factor to determine the temperature for freezing point depression of soil water. We proposed a modified theory to relate T to θ for both unsaturated and saturated conditions. The modified model for the aggregated soil showed good agreement between the calculated and the experimental θ–T relationship. Sensible heat flux decreased due to small temperature gradients in the transition layer between frozen and unfrozen layers. However, when the freezing front advanced further below the soil surface, latent heat of freezing appeared to be larger than the sensible heat because the phase of soil water changed to ice. The heat transport in the transition layer should be taken into account for better prediction of soil-freezing in Andisol.