A new model for predicting thermal conductivity of unsaturated soils using the soil-water characteristic curve

Abstract

Heat and mass transfer processes in porous media, such as soils, strongly depend on thermal conductivity. In contrast to homogeneous materials, the thermal conductivity of soils, especially when they are unsaturated, is highly complex due to the intricate interactions among solid, water, and air phases. Water saturation is one of the most important factors influencing the thermal conductivity. Current models for predicting thermal conductivity, whether empirical, based on mixing theories, or grounded in percolation theory frequently exhibit limitations under varied environmental conditions. To address these challenges, in this study a new model is developed for predicting the thermal conductivity of unsaturated soils, utilizing the Soil-Water Characteristic Curve (SWCC) as a fundamental tool. The proposed approach explicitly links pore-scale thermal conductivity to pore size distribution, subsequently upscaling this relationship to predict normalized thermal conductivity at the macroscale. The model incorporates two parameters, n₁ and η, both of which are strongly related to the pore size distribution. The parameter n₁ is derived from the SWCC while an empirical correlation is suggested between n₁ and η, facilitating practical implementation. The model’s accuracy is validated against a wide range of experimental datasets, demonstrating reliable prediction performance across various soil types and temperature conditions. This model can be effectively used in thermo-hydro-mechanical (THM) coupled modeling for unsaturated soils.