Information Communications Technology,
Civil Engineering Research Section
Recently while various areas throughout Japan have often had quick and unexpected local heavy rain, Hokkaido has also had a lot of the same type of rain. Since rain patterns are not what they used to be, water volume flowing into the dam can suddenly increase even in the cases of rainfall information from gauging and precipitation stations located upsteram of dams, where it is not confirmed.
The Civil Engineering Research Section has conducted research activities for developing the rainfall prediction system for supporting dam management and establishing the anlysis system of runoff suitable for Hokkaido.
The water volume discharged by the dam is decided on the water volume flowing into the dams, which dam management engineers predict based on the data from the Japan Meteorological Agency, gauging and precipitation stations. We hope the prediction accuracy of water volume flowing into the dam will be improved if we predict the rainfall by ourselves and consider the result from the anlysis of how rainfall flows into the rivers, as well.
Although there is a rainfall prediction system in the Japanese Meteorological Agency, the Numerical Weather Forecasting and Analysis System (NuWFAS) developed by Central Research Institute of Electric Power Industry can set models freely. Therefore we can develop a rainfall prediciton system for the next 6 - 72 hours which considers the local characteristics of Hokkaido. Furthermore, we can also establish a runoff anlysis system suitable for Hokkaido by using the Hydrological Evaluation with Evapotranspiration Modeling System (HYDREEMS) which considers the plant evapotranspiration process.
The NuWFAS sometimes cannot predict the front location exactly and consequently a lot of rainfall can be seen in the rivers next to the predicted areas. The HYDREEMS tends to predict larger amounts of river discharge than the actual amount if we directly use the parameters used in areas in Honshu and Kyushu. We will solve these problems and promote R&D activities to aim at a full-scale operation in FY 2014.
Fig.1 Image of the results of NuWFAS
Fig.2 Outline of HYDREEMS