レーダ雨量計を利用したニューラルネットワークによる短期流出計算

Abstract

A surface pluviometer can measure accurate rainfall, but it is difficult to obtain the spatial distribution of rainfall, which is especially important for such a basin as mountainous area, only by ground measurement. The radar rain gauge is not accurate but is good at getting spatial figure of rainfall. To combine these two measuring methods is easy in neural network computation, because it does not require the knowledge of physical relationship between quantities. The results show the improvement of flood estimation at Toyokawa basin compared to the calculation with surface rain data only.