Neural Network Scheme for Radar Rainfall Estimation

Abstract – In the last years, the artificial neural networks (ANN) have proved an attractive approach to non-linear regression problems arising in environmental modeling, such as short-term forecasting of atmospheric pollutant concentrations, rainfall run-off modeling and precipitation now casting using radar, satellite or meteorological data. This neural network scheme enables the network to account for any variability in the relationship between radar measurements and precipitation estimation. This precipitation estimation scheme is a good compromise between the competing demands of accuracy and generalization.