Shanghai is one of the numerous megacities worldwide that experience severe flood events triggered by torrential rainfall. To deal with the undesirable consequences of these events and mitigate the flood hazard, the research of flood reduction measures is necessary. In this effort, the hydrodynamic modelling is a useful tool.
In this master thesis a 2D model was developed for the simulation of urban flood events in Jingan District in the downtown of Shanghai. The main objectives of the thesis were the production of the flood hazard maps for numerous rainfall events and the assessment of the proposed flood mitigation measures. Delft3D Flexible Mesh was used as a tool to produce the inundation maps. Also, several data were considered regarding the grid for the numerical calculation, the surface elevation, the local drainage system and the rainfall events. SOBEK was used for the set up and the preprocessing of the sewer system. For the simulations, data from three historical rainfall events were used: August 2005, August 1997 and September 2013 and five rainfall events with return periods of 1, 3, 5, 10 and 50 years were generated by using the Chicago hydrograph.
For the validation of the model, the rainfall event of August 2005 was used. Although the validation of the model was not proven due to the simplifications that were made in the input data and the lack of data, the model showed that some processes can be simulated, and inundation maps can be produced. By comparing the results that occur with and without the inclusion of the drainage system, it was concluded that the local drainage system should be included in the analysis for the assessment of the flood hazard in an urban area, since its presence plays an important role in the flood reduction. The results showed that the maximum inundation depth can decrease by around 45%. As flood reduction measures, the creation of water storage areas and the increase of the drainage capacity were considered. The water storage areas covered around 10%, or less, of the block areas with available space, leading to a water depth reduction that depends on the location in the map. For the increase of the drainage capacity, the value of 1 m3/s was assigned in a single and in multiple locations in a specific area of investigation. The results showed a percentage of water depth reduction around 15.9% and 45.5%, respectively. For the same location the percentage of water depth reduction due to the water storage areas was 22.5%.
Finally, uncertainties were introduced in the model due to the assumptions and the simplifications that were made in the input data. However, this model can work as a base for future researches to accomplish more realistic results, by improving the current model and adding more updated and precise data.