Neural network prediction of monthly precipitation: Application to summer flood occurrence in two regions of Central Europe

Authors
Bodri, L
Citation
L. Bodri, Neural network prediction of monthly precipitation: Application to summer flood occurrence in two regions of Central Europe, STUD GEOPH, 45(2), 2001, pp. 155-167
Citations number
16
Categorie Soggetti
Earth Sciences
Journal title
STUDIA GEOPHYSICA ET GEODAETICA
ISSN journal
00393169 → ACNP
Volume
45
Issue
2
Year of publication
2001
Pages
155 - 167
Database
ISI
SICI code
0039-3169(2001)45:2<155:NNPOMP>2.0.ZU;2-0
Abstract
Artificial Neural Network (ANN) models were used to forecast precipitation. Three-layer back propagation ANNs were trained with actual monthly precipi tation data from six Czech and four Hungarian meteorological stations for t he period 1961-1998. The predicted amounts are the next monthly precipitati on. Both training and testing ANN results provided a good fit with the actu al data and displayed high feasibility in predicting extreme precipitation.