Skip to Main Content
Table 4

Performance of XAJ and LSTM models for flood forecasting in the training and validation stages

PeriodIndexXAJ modelaLSTM model
Training NSE 0.931 0.978 0.959 0.922 0.883 
KGE 0.894 0.960 0.953 0.914 0.907 
0.965 0.989 0.980 0.961 0.940 
RE (%) −2.309 5.115 −2.783 −4.117 −4.706 
RMSE (m3/s) 72.068 37.220 54.693 76.455 95.704 
Validation NSE 0.891 0.982 0.953 0.901 0.856 
KGE 0.884 0.911 0.924 0.883 0.879 
0.943 0.991 0.976 0.949 0.925 
RE (%) 0.337 5.631 −2.084 −3.091 −5.125 
RMSE (m3/s) 77.346 36.709 55.540 79.448 93.592 
PeriodIndexXAJ modelaLSTM model
Training NSE 0.931 0.978 0.959 0.922 0.883 
KGE 0.894 0.960 0.953 0.914 0.907 
0.965 0.989 0.980 0.961 0.940 
RE (%) −2.309 5.115 −2.783 −4.117 −4.706 
RMSE (m3/s) 72.068 37.220 54.693 76.455 95.704 
Validation NSE 0.891 0.982 0.953 0.901 0.856 
KGE 0.884 0.911 0.924 0.883 0.879 
0.943 0.991 0.976 0.949 0.925 
RE (%) 0.337 5.631 −2.084 −3.091 −5.125 
RMSE (m3/s) 77.346 36.709 55.540 79.448 93.592 

aThe training and validation periods in the XAJ model refer to the flood periods in 2013–2016 and 2017–2019, respectively. Since the flood period in 2012 was taken as the warm-up period of the XAJ model, the data in 2012 were not used for assessing the three models.

Close Modal

or Create an Account

Close Modal
Close Modal