2023 Fiscal Year Research-status Report
Developing ML-based models to estimate flood and sediment response time for tropical re gions
| Project/Area Number |
23KF0258
|
|---|
| Research Institution | Tottori University |
|---|
Principal Investigator |
Haregeweyn N 鳥取大学, 国際乾燥地研究教育機構, 教授 (30754692)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ALEMU DAGNENET 鳥取大学, 国際乾燥地研究教育機構, 外国人特別研究員
|
|---|
| Project Period (FY) |
2023-11-15 – 2026-03-31
|
| Keywords | Peak flow / Lag time / Time of concentration / Machine learning / Sediment concentration |
|---|
| Outline of Annual Research Achievements |
Global Hydrological response time estimation methods were reviewed and archived from 80 published articles to survey and evaluate the accuracy of existing methods across different climatic region. Relationship between catchment size, slope, rainfall intensity, dominant soil texture and land use versus measured time of concentration were analyzed. Additionally, event-based rainfall-peak flow data maintained from our previous study under contrasting climates of Ethiopia were analyzed to determine lag time of peak flows and time of concentration. To support these activities satellite images and laptop computer were purchased.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
The necessary preparations to conduct the field work in Ethiopia in the coming summer are slightly delayed due to security travel restriction to the study sites. Efforts are being made to accelerate the preparation so that the necessary data could be collected as per the initial plan.
|
| Strategy for Future Research Activity |
The future research plan comprises four key activities: 1) Conducting rainy season field surveys in Ethiopia to collect data on peak flow, soil moisture, and sediment concentration for hydrological model validation. 2) Compiling biophysical, hydrological, and meteorological data from various sources to enhance the watershed database. 3) Evaluating the accuracy of existing methods for estimating hydrological response times across different climates. 4) Developing machine learning models to predict flood and sediment response times in tropical regions, enhancing predictive accuracy and efficiency
|
| Causes of Carryover |
Due to unexpected discount for the satellite image being for academic purpose.
We will use the remaining amount to cover travel expenses.
|
