2000 Fiscal Year Final Research Report Summary
Study on real-time flood tovecasting method by distributed runoff model
| Project/Area Number |
11650526
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
水工水理学
|
|---|
| Research Institution | KYOTO UNIVERSITY |
|---|
Principal Investigator |
SHIIBA Michiharu Kyoto University, Faculty of Engineering, Professor, 工学研究科, 教授 (90026352)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TANAKA Kenji Kyoto University, DPRI, Research Associate, 防災研究所, 助手 (30283625)
TACHIKAWA Yasuto Kyoto University, DPRI, Assoc, Professor, 防災研究所, 助教授 (40227088)
HRI Tomoharu Kyoto University, Faculty of Engineering, Assoc. Professor, 工学研究科, 助教授 (20190225)
ICHIKAWA Yutaka Kyoto University, Faculty of Engineering, Research Associate, 工学研究科, 助手 (30293963)
|
|---|
| Project Period (FY) |
1999 – 2000
|
| Keywords | distributed runoff model / real-time flood fore casting / kalman filter / GIS / reservoir operation / meteorology-hydrology combined model |
|---|
| Research Abstract |
1. A real-time flood forecasting algorithm using a distributed runoff model was developed. The frame-work of the algorithm is as follows : 1) an applied river network is divided into several sub-networks, 2) a stochastic distributed runoff model is built by introducing a system noise model with 2-dimensional correlation in a flood routing model applied to each sub-network, 3) Estimates of the state variables are updated by applying a filtering theory to the stochastic distributed runoff model.
2. In general, when calculating the outflow discharge from the slope system by using a distributed runoff model, we require the huge amount of the calculation and large computing time. This fact is a big obstacle in real time river flow forecasting. Then, we developed the lumped runoff model based on a distributed runoff model by which outflow discharge from the slope system was able to be calculated at very high speed.
3. The rainfall forecast with high accuracy is needed for the forecast of the river flow. In this research we assumed to use a meteorology-hydrology combined model as a technique of the rainfall forecast. mind. To improve the accuracy of the meteorology-hydrology combined model, it is necessary to evaluate the soil water content in the surface soil layer correctly. Then we developed the technique by which the soil water content of the surface soil layer was forecasted and corrected by using observed temperature information and the kalman filter theory.
4. When forecasting a flood in real time, we also have to forecast the release discharge from the flood control facilities. Then we developed a system to effectively support the reservoir operation at the flood considering the relation between flood forecasting and release discharge forecasting.
|
