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

Artificial Neural Network (ANN) models were used to forecast precipitation. Three-layer back propagation ANNs were trained with actual monthly precipitation data from six Czech and four Hungarian meteorological stations for the period 1961-1998. The predicted amounts are the next monthly precipitation. Both training and testing ANN results provided a good fit with the actual data and displayed high feasibility in predicting extreme precipitation.