Performance
criteria
ANN
Learning
Test MLR
Learning
Test
  R² 0.95 0.97 0.94 0.96
  E 0.94 0.97 0.94 0.96
  MSE (mm/d)² 0.33 0.16 0.34 0.23
  RMSE (mm/d) 0.57 0.40 0.58 0.48
  MARE (%) 7.03 5.81 7.57 6.97
Where R is the determination coefficient, E is the Nash-Sutcliffe efficiency, MSE is the mean-square error (mm/day) , RMSE is the root mean squared error (mm/ day) and MARE is the mean absolute relative error (%)
Table 5: Comparison between performance criteria of the ANN model with 4 inputs and single neuron and the MLR model.