Real-time radar rainfall forecasts for prediction of urban sewer flow : a case study of Damhusaen Catchment, Denmark | |
| Author | Chhaihort Nai |
| Call Number | AIT Thesis no.WM-20-01 |
| Subject(s) | Rain and rainfall--Denmark--Forecasting Radar meteorology--Denmark--Case studies Hydrodynamic models |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineeing in Water Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | The insufficient reliable quantitative precipitation estimates (QPEs) from radar rainfall is a major challenge to generate accurate short-term quantitative precipitation forecast (QPFs) and to forecast the urban sewer flow and water level. This study develops the different radar gauge adjustment methods to improve the quality of radar QPEs and urban sewer flow forecasting in Damhusåen Catchment (37km2), Copenhagen Denmark. Two commonly used methods from radar-gauge adjustment category (Mean Field Bias -MFB) and geostatistical interpolation (Kriging with External Drift) were applied to adjust the radar rainfall estimate based on gauge network. The adjustment methods will be evaluated based on the flow prediction performance from hydrodynamic model (Mike Urban) and net rainfall volume comparison. The most reliable adjustment method will be applied on short-term quantitative precipitation forecast (QPF) with up to 3h forecast lead time (radar nowcast). The best QPF will be input into hydrodynamic model for flow and water level forecast with different lead time. The results indicate that the dynamic MFB method was proved to be useful method to improve the quality of radar rainfall estimate and have the advantage to adjust in real time application. Moreover, Kriging with external drift is the radar-gauge merging method have shown the potential to capture the spatial resolution of radar field but fail to generate the high quality QPF as this method damage the continuity of spatial structure between consecutive rainfall fields. Thus, dynamic MFB with 6h time moving window was selected as the most reliable method in this study with the consistency performance to minimize the volume and peak error in flow prediction. The flow and water level forecast quality are degraded in function to length of lead time, yet with 3h forecast lead time, the forecast performance still found to be in reliable state. |
| Year | 2020 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Babel, Mukand Singh; |
| Examination Committee(s) | Sundaram, S. Mohana;Sutat Weesakul;Virdis, Salvatore G.P.;Madsen, Henrik; |
| Scholarship Donor(s) | Loom Nam Khong Pijai (Greater Mekong Subregion) Scholarships; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2000 |