2023 Fiscal Year Annual Research Report
Development of high resolution global-flood forecasting system with long lead time
| Project/Area Number |
21K14386
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
馬 文超 東京大学, 生産技術研究所, 特任研究員 (60743101)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | ensemble member |
|---|
| Outline of Annual Research Achievements |
A real-time flood forecasting system based on ECMWF multi-ensembles has been established. The multi-ensembles results can provide more reliable forecasting results by considering the ensemble values as uncertain. The historical run, from 2001 to 2020, was conducted as a benchmark for estimating the database for flood risk. Furthermore, based on the established database, we calculated the global flood risk. The results imply that some regions have a higher risk, which should be validated by observation for a historical period before the results are shown. This task has been targeted by searching leading methods, which is still not fixed since data availability and consistency are not optimal, e.g., remote sensing data. However, because the resolution of cama-flood was set as 0.25 degrees, treatment of observation data is needed. Although the global-scale flood forecasting system has been finished for the current stage, more data analysis is necessary, which will be done in further work.
Further, tasks still need to be completed regarding the real-time system construction. This is because accessing and downloading the ECMWF-forcing data takes time. The code for decoding the multi-ensemble members forcing data must be improved.
|
