The objective of this study was to evaluate the performance of different mathematical models to adjust the humidity data as a function of the apparent dielectric constant of the soil and to evaluate the feasibility of the use of wave guides of two and three rods with different spacings. Disturbed soil samples were packed in PVC tube segments of 0.075 m diameter. Two sets of 24 have guides were constructed. One of this sets had a capacitor. In each set one half of the wave guides had two rods and the other half contained three rods. The rod spacing varied from 0.009 to 0.022 m. Soil water content data from gravimetry and soil bulk dielectric constant values from Trase System analyzer were collected during drying process with water content values ranging from 0.31 to 0.13 m³ m-3. Five mathematical models were fitted to water content and bulk dielectric constant data. The Malicki's model was the most adequate for estimating soil water content as a function of bulk dielectric constant. The wave guides with three rods 0.017 m apart from each other showed the best performance. The three-rod wave-guides without capacitor performed better for water content determination than the two-rod wave-guides without capacitor. The three-rod wave-guides without capacitor performed better than three-rod wave-guides with capacitor.
calibration; wave guide; soil water content; time-domain-reflectometry
