Calibração do sensor de umidade do solo modelo YL-69 para manejo da irrigação

Abstract

Irrigated agriculture contributes to food production, however, it consumes a huge amount of water. In irrigation, the soil moisture is one of the parameters that determines its efficiency allowing the optimization of water and energy. Therefore, the objective was to calibrate soil moisture sensors YL-69 to obtain calibration equations in four reference soils of the State of Pernambuco and to compare its accuracy with two ECH2O sensors in the estimation of soil moisture. Deformed and non-deformed soil samples were collected in the 0-0.40 m depth layer. The deformed samples were sifted in 4.75 mm mesh, stored in pots of 4.66 L maintaining the soil bulk density (1.29 Mg m-3 – Neossolo Regolítico, 1.45 Mg m-3 – Latossolo Amarelo, 1.54 Mg m-3 – Espodossolo Humilúvico and 1.23 Mg m-3 Argissolo Vermelho-Amarelo). For each soil, 5 pots were drilled in the base, filled with 5 cm of gravel and covered with a cover to prevent soil loss during drainage. An YL-69 sensor was installed in each vessel 3.5 cm deep. The sets (pot + gravel + bidim + soil + sensor) were weighed daily and the soil moisture readings were collected with the aid of an Arduino board. The weighings and the readings recorded by the sensors were carried out until the humidity variation of the soil mass became constant. There was a good correlation between the readings of measured volumetric moisture and the one estimated by the YL-69 sensor (R2 > 0.90 for all soils). The calibration results indicated that factory calibration errors can be reduced obtaining RMSE values  0.02 m3 m-3 for all soils. After calibration, the soil moisture measurements by the YL-69 sensor and the ECH2O probes in the studied soils were compatible. It was concluded that the soil moisture sensor YL-69 should be subjected to a specific calibration for the soil in question and could even replace the ECH2O probes in the humidity estimation from soil.