Influences of root fragments on soil thermal property measurements with heat pulse method

Abstract

Heat pulse (HP) method has been widely applied for measuring thermal properties, heat fluxes, and soil–water evaporation for near-surface soil layers. In topsoil layers, the presence of roots violates the underlying assumption of HP method that the soil under test is homogeneous. In this study, numerical and laboratory experiments were conducted to test the influence of root fragments on soil thermal property measurement with HP sensors. Soybean root fragments, with a water content of about 0.74 m3 m−3 and diameters ranging from 0.5 to 4 mm, were carefully repacked into soil samples that were used for HP measurements. Results showed that the effects of roots on soil thermal property measurements depended on root size, root location and directions relative to the heater probe, and the thermal property differences between soil sample and root. Roots showed negligible influence on heat pulse data when the root fragments were positioned outside the HP measurement region (more than probe space away from heater probe), or root diameters (d) were less than 1 mm, or in oven-dried soil samples. On relatively wet soils (i.e., with water content of 0.05–0.25 m3 m−3), when root fragments were located in between the heater and sensing probes (parallel to the probe planes), (1) the heat capacity (C) and thermal conductivity (λ) differences between root-free soil and root-soil mixture increased proportionally with d increased from 1 mm to 4 mm; (2) the differences in absolute values of C and λ between the root-free soil and soil-root mixture were highly dependent on soil water content. The errors in soil C and λ estimates were larger when the root fragment was parallel with than perpendicular to or inclined 45 degrees to the probe plane, and the error overall decreased with a smaller ρb and sand fraction. When the root fragments were positioned within the HP measurement region, or d were larger than 1 mm, or in relatively wet soil samples, it is important to consider root effect when interpreting HP signals.