Skip to Main Content
Table 4

Comparison of different performance measures of ANN modelling for the 1-day-ahead streamflow forecasting

Rain gauge network and ANN models usedTraining phase
Validation phase
Testing phase
NRMSEMAENSCECCNRMSEMAENSCECCNRMSEMAENSCECC
Current rain gauge network (BoM's base network): ANN model-1 0.251 2.236 0.937 0.968 0.296 1.511 0.913 0.955 0.284 0.946 0.919 0.961 
Optimal rain gauge network considering no additional fictitious stations: ANN model-2 0.190 1.620 0.964 0.982 0.248 1.327 0.939 0.969 0.264 0.923 0.930 0.969 
Augmented optimal rain gauge network considering additional fictitious stations: ANN model-3 0.183 1.425 0.967 0.983 0.250 1.115 0.938 0.968 0.232 0.658 0.946 0.974 
Rain gauge network and ANN models usedTraining phase
Validation phase
Testing phase
NRMSEMAENSCECCNRMSEMAENSCECCNRMSEMAENSCECC
Current rain gauge network (BoM's base network): ANN model-1 0.251 2.236 0.937 0.968 0.296 1.511 0.913 0.955 0.284 0.946 0.919 0.961 
Optimal rain gauge network considering no additional fictitious stations: ANN model-2 0.190 1.620 0.964 0.982 0.248 1.327 0.939 0.969 0.264 0.923 0.930 0.969 
Augmented optimal rain gauge network considering additional fictitious stations: ANN model-3 0.183 1.425 0.967 0.983 0.250 1.115 0.938 0.968 0.232 0.658 0.946 0.974 

NRMSE, normalized root mean squared error; MAE, mean absolute error; NSCE, Nash–Sutcliffe coefficient of efficiency; CC, correlation coefficient.

Close Modal

or Create an Account

Close Modal
Close Modal