Despite the several existing initiatives to obtain evapotranspiration data for several scientific analysis, it remains a challenge when dealing with data scarcity, especially when it is coupled with many data gaps of long-time climatological data series. In this study, we developed two Python scripts to perform reference evapotranspiration calculations in data scarcity conditions, filling potential data gaps, and testing them in the Urucuia River watershed.The first script (ETO_calc.py) is able to calculate evapotranspiration both by the Penman-Monteith method, adapting it according to user data availability, and by the Hargreaves-Samani method. The second script (Filling_gaps.py) fills out potential data gaps based on input from neighboring stations and tests the adequacy of the filling. The first script was tested by comparing the data generated by it with the online calculator 'ETCalc', while the second was tested by filling in a station's data from five neighboring stations, comparing the original data with the filled data. The output of the first script was practically identical to the online calculator used for comparison, which confirms the adequacy of the estimation made by ETO_calc.py. As for the second script, the regional weighting method and both linear and multiple regression methods were the methods that best filled in the missing data for the main station. This is due to their better fit to the historical variation of the data compared to the other tested methods. The results allow us to conclude that both scripts proved to be suitable tools for estimating evapotranspiration data.