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

Environmental Sciences Europe

Table 6 Mean absolute error of the prediction for the different hydrological phases of the record-breaking 2006 flood and the flood in 2013

Day	Required precision (cm)	MAE (cm) for the phases of the 2006 flood			MAE (cm) for the phases of the 2013 flood
Day	Required precision (cm)	Rising	Peak	Falling	Rising	Peak	Falling
First	± 5	6.2	2.2	3.4	6.2	2.2	2.1
Third	± 15	17.6	9.7	13.2	18.2	9.3	5.0
Fifth	± 25	39.2	31.1	21.2	43.0	15.4	12.3
Seventh	± 35	70.6	55.3	34.3	78.7	21.3	30.3