Forecast of Reservoir Flooding in the Yellow River, China (PAP016656)

Código

PAP016656

Autores

M. Castro-Gama, I. Popescu, A. Mynett, L. Shengyang

Tema

Flood forecasting and early warning systems

Resumo

During the last 50 years, China has developed economically at a steady rate. Economical development and population increase has triggered extensive use of water resources. At the same time climate change increases the stress to update reservoir operation strategies to guarantee proper flood management. The Yellow river is an example of Chinese water resources management under these circumstances. In this case, the Yellow River Conservancy Commission (YRCC) operates a multi-reservoir system dealing with flood and sediment management along its reaches. A methodology based on Multivariate Regression (MR) and Generalized Regression Neural Networks (GRNN) is applied. Predictor variables are related to reservoir operation (e.g. maximum season operative level in the reservoir, initial level of the reservoir). Dependant variables are related to flood hydrographs variables (e.g. flood volume, peak discharge, time of peak discharge). It is shown that it is possible to perform statistical inference in a highly intervened stream and reservoir driven system such as Yellow River, under a climate change scenario for extreme flood events forecast.