Manejo de irrigação via sensores de solo e fontes nitrogenadas no crescimento da cana-de-açúcar

Abstract

The management of irrigation, through the use of electronic sensors that estimate in real time the soil water content, and of N-fertilizer have demonstrated a positive synergistic effect on plant development and productive potential of different varieties of sugarcane. Thus, the objective of this research was to evaluate the performance of different electronic sensors of soil moisture in the optimization of irrigation management, aiming to increase the precision in the rational use of water, as well as to compare the initial development of sugarcane under different sources of nitrogen. The experiment was carried out in a protected environment at the Agricultural Engineering Department of the Federal Rural University of Pernambuco, campus Recife-PE, between April and May 2019, initially for the calibration of sensors; and, from December 2019 and March 2020, to assess their performance in the initial development of the sugarcane under different sources of nitrogen. For calibration, the EC-5, 5TE and Diviner sensors were positioned vertically in the pots, with four replicates for each type of soil (sandy clay loam and sandy textural classes). After reaching capillarity saturation, the pots were placed on a bench in order to allow the drainage of excess water. Then, the pots were weighed and the respective soil moisture readings were performed, representing the moisture equivalent to the field capacity. Weighing and reading each electronic sensor was performed daily and at a fixed time (8 am). The results were subjected to regression analysis and the statistical indexes Willmott, RMSE, determination coefficient and mean error were used to assess the quality of the fit between the measured values (gravimetric moisture) and estimated by soil moisture sensors. The sensors presented linear model calibration equations for both soils, and the analyzes showed a good correlation between the readings measured volumetric moisture and the estimated for both soils with R2 above 0.94. The manufacturers calibration equation of all sensors, EC-5, 5TE and Diviner overestimated 2.8; 13.86 and 7.51%, respectively, the real moisture values for the sandy soil and underestimated by 21.88, 15.63 and 7.64%, respectively, in the sandy clay loam soil, resulting in errors in determining the irrigation depth and, consequently, in the amount of water required by the crop. Thus, and considering the importance of measuring soil moisture for the management of irrigated agriculture, the results of this research reinforce the need to calibrate the capacitive sensors ECH2O and Diviner in sandy clay loam and sandy soils for irrigation purposes, aiming at the use rational use of natural resources. After this step, the second experiment was carried out to evaluate the performance of soil moisture sensors (EC-5, 5TE and Diviner 2000) and the effects of application of two sources of nitrogen ((NH4)2SO4 and NH4NO3), in the biometric variables. Therefore, the treatments were arranged in a completely randomized design, in a 3 x 2 factorial scheme, with 4 replications, totaling 24 experimental plots. The irrigation was carried out automatically, using an electronic controller, in order to apply the irrigation depths according to the established treatments. Biometric variables were assessed monthly and at the end of the experiment. The results of the biometric variables were subjected to analysis of variance, at the level of 0.05 probability. As a result for the sandy clay loam soil, the ammonium nitrate treatments using the 5TE sensor and ammonium sulfate with EC-5 showed greater water use efficiency for dry mass 0.17 and 0.19 kg m-3, respectively, and fresh mass with 0.49 kg m-3 and for both treatments.