Comparison of short-term rainfall forecasts for prediction of sewer flow in urban area: a case study of Damhusaen catchment, Copenhagen | |
| Author | Thol, Thaileng |
| Call Number | AIT Thesis no.WM-19-02 |
| Subject(s) | Radar meteorology--Measurement Rain and rainfall--Forecasting Rain and rainfall--Denmark--Copenhagen--Measurement |
| Note | A thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering in Water Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | Climate change and growing urbanization increase the pressure on the drainage systems. For real-time control of the drainage system and flood prediction, the spatial predictability in both time and space, as well as the rainfall intensities, are very important. As a function of lead time, the uncertainty will rapidly increase to a point where skill is lost and the nowcasts can no longer be applied for real-time control. The study aims to compare the potential of different rainfall product as well as the short-term rainfall forecasting for estimation of the flow in Damhusåen catchment, Copenhagen. In addition, the efficiency of data assimilation also applied to test its performance on flow forecasting from radar nowcast. A significant improvement can be found with the flow simulation from the corrected radar data by applying mean field bias adjustment. The radar data could extend the reliability of rainfall data with better spatial variability of rainfall by merging rain gauge data. From the flow forecasting results of 2 storm events in this study, the flow forecast form radar has the performance of forecasting for lead time up to 2-3 hours depending on the storm type and magnitude. The flow forecasting from NWP as input data could provide a longer time horizon of forecasting than radar extrapolation yet need to be compromising on both temporal and spatial resolution. In this study, the forecasted flow at the Inlet to WWTP can be extended for lead time up to 12 hours. The data assimilation technique which was applied for the flow forecasting from the radar nowcast could improve the forecasting result. The volume error can be reduced to 13% at the assimilation location and to 3% at the verification location. These results suggest that combining the three forecast products might provide the best result. If the processing time of Radar nowcast could be decreased, this could be combined with the NWP nowcast for the first approx. 2 hours of the forecast and improve the forecast performance. The data assimilation also could reduce the uncertainty of the model during the hindcast period, to make the system and model have the same state at the time of initiation of forecasting. |
| Year | 2019 |
| 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) | Sutat Weesakul;Virdis, Salvatore G.P.;Madsen, Henrik;Mark, Ole; |
| Scholarship Donor(s) | Deutscher Akademischer Austausch Dienst (DAAD), Germany;Asian Institute of Technology Fellowship ; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2019 |