Precipitation prediction with neural networks

Research output: Article

2 Citations (Scopus)

Abstract

Dramatic floods occurred in Central Europe in recent summers, Hungary having been seriously affected in its eastern part. Predictive approach based on modeling flood recurrence may be helpful in flood management. Summer floods are typically characterized by saturated catchment due to long-lasting heavy precipitation followed by a sudden extreme rainfall. In present work, an artificial neural network (ANN) models were evaluated for precipitation forecasting. A back propagation neural networks were trained with actual annual and monthly precipitation data from east Hungarian meteorological stations for a time period of 38 years. Predicted amounts are next-year-precipitation and summer precipitation in the next year. The ANN models provided a good fit with the actual data, and have shown a high feasibility in prediction of extreme precipitation.

Original languageEnglish
Pages (from-to)207-216
Number of pages10
JournalActa Geodaetica et Geophysica Hungarica
Volume36
Issue number2
DOIs
Publication statusPublished - aug. 16 2001

ASJC Scopus subject areas

  • Building and Construction
  • Geophysics
  • Geology

Fingerprint Dive into the research topics of 'Precipitation prediction with neural networks'. Together they form a unique fingerprint.

  • Cite this