Forecasting monthly pan evaporation using hybrid additive regression and data-driven models in a semi-arid environment

Abstract Exact estimation of evaporation rates is very important in a proper planning and efficient operation water resources projects agricultural activities. Evaporation affected by many driving forces characterized nonlinearity, non-stationary, stochasticity. Such factors clearly hinder setting up rigorous predictive models. This study evaluates the predictability coupling additive regression model (AR) with four ensemble machine-learning algorithms—random Subspace (RSS), M5 pruned (M5P), reduced error pruning tree (REPTree), bagging for estimating pan rates. Meteorological data encompass maximum temperature, minimum mean relative humidity, wind speed from three different agroclimatic stations Iraq (i.e., Baghdad, Mosul, Basrah) were utilized as predictor parameters. The addition to sensitivity analysis was employed identify best-input combinations evaluated methods. It demonstrated that AR-M5P estimated higher accuracy than others when combining speed, temperatures input provided best performance indicators, i.e., MAE = 33.82, RMSE 45.05, RAE 24.75, RRSE 28.50, r 0.972 Baghdad; 25.82, 35.95, 23.75, 29.64, 0.956 Mosul station, respectively. outcomes this proved superior hybridized methods addressing such intricate hydrological relationships hence could be other environmental problems.