A large-scale waterlogging investigation in a megacity

With a growing number of waterlogging events occurring in large cities, a better understanding and prediction of the urban flooding processes are essential in order to reduce their impacts. A two-dimensional hydrodynamic model was used to simulate a surface water flooding process due to a heavy rainstorm event in the urban area of Beijing. Through incorporating road networks over the large area, an enhanced digital elevation model was presented to describe overland flow through surface pathways and waterlogging forming along the main roads. Rainfall data obtained from a large number of rain gauge stations in Beijing City throughout the whole storm event were acquired. The model was capable of reproducing the hydrodynamic process at the city scale, and it was found that surface runoff generated quickly from the surrounding community areas to roads and accumulated in depression areas. Waterlogging sites reported in the media were extracted, with most of the sites being on roads and predicted reasonably. Waterlogging issues were found generally to be a local feature and closely associated with rainfall intensity.