Author(s)

X. Vargas, R. González, P. Valck & C. Núñez

Abstract

A flood warning system using meteorological predictions coupled to a neural network and transfer function hydrological model X. Vargas, R. Gonzblez, P. Valck & C. Ntiiiez Civil Engineering Department, Universidad de Chile Santiago, Chile Abstract In Chile, due to very steep mountainous areas near urban zones, it is almost impossible to establish forecasting hydrological models with enough lead-time to take non-structural measures to mitigate the effects of floods. During construction of a highway inside the Mapocho River, a meteorological model was performed to forecast the magnitude of rainfall and temperature three days in advance, making possible an improvement in the lead-time of predictions. This paper analyses the use of these results in a basin, near Santiago, the country's capital city, which is characterized by having a variable snowfall area depending on prevailing temperature. Two gage stations situated in tributaries, located a certain distance upstream from the highway works, cont

Keywords