Performance evaluation of a recently developed soil water content, dielectric permittivity, and bulk electrical conductivity electromagnetic sensor

CS655 Water Content Reflectometer (WCR) (Campbell Scientific, Inc., Logan, UT, USA) is a recently developed multiparameter electromagnetic sensor measuring volumetric soil water content (theta), bulk soil electrical conductivity (sigma(b)), and temperature. In this study, CS655 sensor's performance was investigated through laboratory experiments using a set of liquids with known dielectric and salinity properties and soils featuring a wide range of conditions (including physico-chemical properties, water regimes and salinity of the water solution). The sensor's performance was analyzed in comparison with WET and ML2 theta probe (Delta-T Device Ltd, Cambridge, UK) dielectric sensors. It was found that CS655 apparent dielectric permittivity (epsilon(a)) readings were higher than that of Topp's permittivity-water content relationship, especially at higher soil water content values, as much as 12 dimensionless permittivity units. The results suggested that the relationship between experimentally measured soil water content (theta(m)) root epsilon(a) and ea was strongly linear (0.911 < R-2 < 0.997). The most accurate results were provided by the multi-point calibration method. Moreover, for the case of non-conductive soils, Kelleners's method provides similar epsilon(a) values with CS655 sensor readings for a maximum sigma(b) value of about 0.8 dS m(-1). Concerning the use of empirical calibration equations, which include the effect of bulk electrical conductivity in epsilon(a), the obtained results suggest that the equation theta = a root epsilon(a)+ b + c sigma(b) has a strong physical basis and adequate performance (RMSE < 0.03 m(3) m(-3)). Finally, the investigation of the epsilon(a)-sigma(b) relationship revealed that it is strongly linear and that slope mainly depends on the pore water electrical conductivity (sigma(p)), which is closely related to the soil solution salinity in contact with plant roots that influences plants growth. The Malliki-Walzcak model provide more reliable sigma(p) predictions as compared to the Hilhorst model.