Measuring Near-Surface Soil Thermal Properties with the Heat-Pulse Method: Correction of Ambient Temperature and Soil-Air Interface Effects

Large temperature gradients and proximity to the soil-air interface affect the measurement accuracy of thermal properties with the heat-pulse probe (HPP). The objective of this study is to improve the HPP methodology for measuring soil heat capacity (C) and thermal conductivity (l) in the surface soil layer by reducing the effects of ambient temperature variation and the soil-air interface. Thermal properties of eight soil layers (0-6, 2-6, 6-12, 12-18, 18-24, 24-30, 30-36, and 36-42 mm) were measured with the HPP in a loamy sand soil in two experiments with varying moisture conditions. Results were compared with values estimated with the de Vries models. Temperature drift caused by natural warming and cooling was removed from the observed trend of ambient temperature change with time. The influence of the soil-air interface was taken into account with a new solution using pulsed infinite line source theory with adiabatic boundary conditions. When ambient temperature change was considered, the HPP were capable of providing reasonable thermal property results at soil depths greater than 6 mm. For the 0- to 6-mm depth layer, the HPP signals were affected simultaneously by ambient temperature drift and the soil-air interface, and mixed scenarios were produced. Around midday, the HPP temperature by time curves were distorted so seriously that the effect of ambient temperature drift could not be removed. At other times, large errors were observed in the estimated C and l data, even when ambient temperature drift was removed. When the effect of the soil-air interface was accounted for, however, C and l measurement accuracies in the 0- to 6-mm layer were improved significantly. The RMSE of C was reduced from 0.43 MJ m(-3) degrees C-1 to 0.29 MJ m(-3) degrees C-1 in Exp. 1, and from 0.52 MJ m(-3) degrees C-1 to 0.34 MJ m(-3) degrees C-1 in Exp. 2. Significant error reduction was also observed in l measurements. Thus, to obtain accurate field measurements of shallow soil thermal properties, the ambient temperature drift and the impact of the soil-air interface must be properly taken into account.