Water resource management and sustainable agriculture depend on the precise measurement of reference evapotranspiration ‘ET0’. Four regression models’ performances are assessed in this study: Bayesian Ridge Regression ‘BRR’, Gaussian Process Regressor ‘GPR’, Random Forest with Quantile Regression ‘RFQR’, and Gradient Boosting Regressor ‘GBR’. And their ability to predict ET0 while incorporating prediction interval analysis to measure uncertainty is investigated. The models were assessed using performance indicators such Coefficient of determination ‘R2’, Mean Absolute Error ‘MAE’, Root Mean Squared Error ‘RMSE’, Mean Squared Error ‘MSE’, Mean Prediction Interval ‘MPIW’, and Prediction Interval Coverage Probability ‘PICP’. The results show that GPR is the most consistent model, exhibiting remarkable accuracy and the smallest prediction intervals, proving its capacity to manage the intricacy of ET0 data. While BRR was computationally efficient, it had trouble with wider prediction intervals. In contrast, GBR and RFQR demonstrated competitive performance, striking a balance between accuracy and uncertainty. This study provides a strong foundation for future research and applications in a variety of climates by highlighting the complementary nature of uncertainty quantification and prediction precision.