Knowledge of evapotranspiration at various stages of plant development allows irrigation management to be more efficiently carried out according to the actual crop demand. There are a variety of methods that can be used to determine reference evapotranspiration, and due to the high variability of parameters influencing this phenomenon, it is common to have studies that compare different methods to evaluate which one has the greatest applicability to the study location. Monthly climatological data from the National Institute of Meteorology - INMET were collected for the municipality of Ilhéus/BA. Reference evapotranspiration was estimated using the Thornthwaite, Camargo, and Priestley & Taylor methods, and their results were compared with the standard Penman-Monteith method (FAO-56) using statistical indices, the Wilmott agreement index, Pearson correlation coefficient, and confidence index. The Priestley-Taylor method was classified as "Excellent" (c > 0.85). The Thornthwaite and Camargo methodologies were classified as "Fair" (0.61 < c < 0.65) and "Good" (0.66 < c < 0.75), respectively. The Priestley-Taylor method overestimated the values of the PMFAO method, but obtained a higher coefficient of determination with values close to those found by the standard method, and can be used as an alternative to the Penman-Monteith (FAO-56) reference method for the Ilhéus-BA region. The Thornthwaite and Camargo methods had low coefficients of determination and obtained the worst statistical performance, and therefore should be used with caution.