A Theoretical model of the thermal conductivity of idealized soil

Accurate prediction of soil thermal conductivity is of prime importance in the numerical simulation of heat transmission through soils. This paper focuses upon empirical and semiempirical prediction methods for soil thermal conductivity. A family of empirical correlations are presented which relate soil thermal conductivity to saturation for five soil types: gravel, sand, silt, clay, and peat. These correlations are developed from a database of measured data available in the literature. Also, a theoretical model of soil thermal conductivity is developed for granular materials composed of rotund particles in an almost dry state. This theoretical model includes the effects of the micro-structure and the conductivity of the solid phase. It explicitly relates soil thermal conductivity to dry density and agrees well with experimental data. This paper also presents a review and discussion of those factors which affect soil thermal conductivity, previously reported prediction methods, and conductivity measurement techniques.