Improved Prediction of Monthly Pan Evaporation Utilising Support Vector Machine Technique

Evaporation is a key element for irrigation system design, water resource management, and hydrological modelling. In this research work, monthly evaporation (Ep) was projected by utilising Support Vector Machine (SVM). Monthly meteorological statistics from a Malaysian weather station were utilised for training and testing the model by employing climatic aspects, such as mean temperature, minimum temperature, maximum temperature, wind speed, relative humidity, and solar radiation for the period 2000 to 2019. Various models were formulated by utilising diverse combination of input elements and other model parameters. The performance of the formulated model was assessed by utilising standard statistical indices. The outcomes indicated that the developed SVM model can significantly improve the accuracy of monthly Ep projections. The achieved performance measures are, $\mathrm{R}^{2}=0.970$, MAE=0.067, MSE=0.007, RMSE=0.087, RAE=0.163 and RSE=0.029.