Skip to main content

Table 7 Proportion (%) of the modelled data relative to the required precision intervals for the prediction of low and medium stages obtained using the LSTM model and DLCM for the validation period (2014–2019)

From: Water level prediction using long short-term memory neural network model for a lowland river: a case study on the Tisza River, Central Europe

Day

Req. prec. (cm)

Low stages (< 240 cm)

Medium stages (240‒649 cm)

LSTM model

DLCM

LSTM model

DLCM

Inside

Lower

Higher

Inside

Lower

Higher

Inside

Lower

Higher

Inside

Lower

Higher

First

 ± 5

78.9

13.1

8.0

70.4

16.3

8.9

59.2

13.7

27.1

39.1

28.1

29.3

Third

 ± 15

84.2

11.2

4.6

76.8

15.1

7.4

62.6

11.7

25.7

45.5

24.0

28.9

Fifth

 ± 25

82.7

13.2

4.1

80.2

13.8

5.4

53.8

18.3

29.1

48.2

22.0

29.1