PERFORMANCE EVALUATION OF REFERENCE POTENTIAL EVAPOTRANSPIRATION ESTIMATION METHODS FOR THE CITY OF BENJAMIN CONSTANT, AMAZONAS

Abstract

Water availability for a crop can be best defined by the range in which the climate will allow the plant to maintain a transpiration rate equal to the rate of water absorption by the roots.

Thus, the objective was to evaluate the performance of the indirect reference evapotranspiration estimation methods of Blaney-Criddle, Camargo, Hargreaves-Samani, Jensen-Haise, Thornthwaite, Makkink, FAO 54 da Radiação and Blaney-Criddle-Frevert and with the standard method by Penman-Monteith. The meteorological data used were obtained from the Benjamin Constant conventional meteorological station of the National Institute of Meteorology, comprising monthly data that were the 10-year provisional normal. The statistical indicators used were the correlation coefficient, determination coefficient, accuracy index and the safety or performance coefficient. The methods adjusted to the Penman-Montheith Standard with “optimal” performance were Makkink, FAO 54 Radiation, Jensen-Haise, Hargreaves-Samani, Blaney-Criddle-Frevert and Blaney-Criddlle, recommended for the municipality of Benjamin Constant, AM. Camargo's model showed “poor” performance and can be recommended after local adjustments. And the Thornthwaite model performed “poorly”, and cannot be recommended due to its low accuracy and precision in relation to the standard method.