PERFORMANCE OF MODELS TO ESTIMATE THE REFERENCE EVAPOTRANSPIRATION IN THE TOMÉ AÇÚ-PA MICROREGION

The estimation of reference evapotranspiration (ETo) is a parameter that allows the management of agricultural activities in a region and the rational use of water resources, enabling the calculation of water volume to be applied. This study aimed to evaluate the performance of empirical methods to calculate the reference evapotranspiration in the microregion of Tomé-Açú, in the state of Pará, Brazil. The performance of the simplified models was verified through simple linear regression, using the statistical indicators Mean Bias Error (MBE), Relative Mean Bias Error (rMBE), Root Mean Square Error (RMSE), and Relative Root Mean Square Error (rRMSE). The models that presented lower data dispersion were Jensen-Haise, Hargreaves-Samani, Turc, Makkink, and Priestley-Taylor, compared to the standard equation of Penman-Monteith. When analyzing the accuracy of the models, the models of Hargreaves-Samani, Makkink, and Camargo were classified as good. The models of Linacre, Blaney-Criddle, and e Jensen-Haise are not recommended to the microregion of Tomé-Açu. In the lack of the necessary data to estimate the reference evapotranspiration by the standard method of Penman-Monteith-FAO 56, the equation that obtained the best adjust was the model of Hargreaves-Samani. Thus, it may be applied for the studied region.