Models for moisture estimation in different horizons of yellow argisol using TDR

The determination of soil moisture is very important because it is the property with the most influence on the dielectric constant of the medium. Time-domain reflectometry (TDR) is an indirect technique used to estimate the water content of the soil (?) based on its dielectric constant (Ka). Like any other technique, it has advantages and disadvantages. Among the major disadvantages is the need for calibration, which requires consideration of the soil characteristics. This study aimed to perform the calibration of a TDR100 device to estimate the volumetric water content of four horizons of a Yellow Argisol. Calibration was performed under laboratory conditions using disturbed soil samples contained in PVC columns. The three rods of the handcrafted probes were vertically installed in the soil columns. Weight measurements with digital scales and daily readings of the dielectric constant with the TDR device were taken. For all soil horizons evaluated, the best fits between the dielectric constant and the volumetric water content were related to the cubic polynomial model. The Ledieu model overestimated by approximately 68 % the volumetric water content in the A and AB horizons, and underestimating by 69 % in Bt2, in relation to volumetric water content obtained by gravimetry. The underestimation by linear, Topp, Roth, and Malicki models ranged from 50 % to 85 % for all horizons.