An evaluation of soil thermal conductivity models based on the porosity and degree of saturation and a proposal of a new improved model

Abstract

Soil thermal conductivity is a key factor for the soil heat transfer analysis, and is widely applied in many fields. However, it is difficult to measure the thermal conductivities of various soils with different porosities and degrees of saturation. Therefore, an accurate soil thermal conductivity model is essential for practical engineering. In this research, the comparisons and evaluations of eight soil thermal conductivity models (including seven from other scholars and a new improved model) are conducted based on the measured data from published literature. Additionally, these models are categorized into two groups: theoretical models and empirical models. The results show that the predictive capabilities of the empirical models are higher than those of the theoretical models because the first use fitted values of the empirical parameters better fit the experimental data. However, the theoretical models have definite physical meanings. The new proposed model shows better performance, and its Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values when the thermal conductivities of seven soils are calculated are less than 0.1 Wm−1K−1. This work provides references for the selection and application of soil thermal conductivity models