A generalized model for estimating effective soil thermal conductivity based on the Kasubuchi algorithm

Abstract The effective soil thermal conductivity ( λ eff , W·m −1 ·K −1 ) has essential applications in disciplines such as earth and environmental science, agriculture and engineering. A plenty of models are available in the literature to predict λ eff , but most of them are restricted to specific soil types or water content ranges. The objective of this study was to develop a generalized thermal conductivity model based on the Kasubuchi model for accurate applications in soils of various textures from dry to saturation. Soil pedotransfer function (PTF) has been established to calculate model parameters based on easily available soil physical properties. The performance of the new model was evaluated and compared with the original Kasubuchi model and six other widely applied models. Twenty-two soil samples of various textures and water contents were used to calibrate and validate the new model. The results showed that the modified model performed best among these investigated models (RMSE = 0.124 W·m −1 ·K −1 , NSE = 0.968). This modified model is promising for wide applications with accurate estimation of λ eff .