Artificial Neural Network (ANN)-Based Long-Term Streamflow Forecasting Models Using Climate Indices for Three Tributaries of Goulburn River, Australia

Water resources systems planning, and control are significantly influenced by streamflow forecasting. The in northern north-central regions of Victoria (Australia) is different climate indices, such as El Niño Southern Oscillation, Interdecadal Pacific Decadal Indian Ocean Dipole. This paper presents the development ANN model using machine learning with multi-layer perceptron Levenberg algorithm for long-term forecasting three tributaries Goulburn River located within through establishing relationships between indices streamflow. were used input predictors models’ performances analyzed best fit correlation. higher correlation values developed models evident from Pearson regression (R) ranging 0.61 to 0.95 reveal acceptability. accuracies evaluated statistical measures Root Mean Square Error (RMSE), Absolute (MAE) Percentage (MAPE). It found that considering R, RMSE, MAE MAPE values, ENSO has more influence (61% 95%) on than other drivers. Moreover, it concluded Acheron can be confidently forecast even six-months ahead.