A Framework on Fast Mapping of Urban Flood Based on a Multi-Objective Random Forest Model

Fast and accurate prediction of urban flood is of considerable practical importance to mitigate the effects of frequent flood disasters in advance.To improve urban flood prediction efficiency and accuracy,we proposed a framework for fast mapping of urban flood:a coupled model based on physical mechanisms was first constructed,a rainfall-inundation database was generated,and a hybrid flood mapping model was finally proposed using the multi-objective random forest(MORF)method.The results show that the coupled model had good reliability in modelling urban flood,and 48 rainfall-inundation scenarios were then specified.The proposed hybrid MORF model in the framework also demonstrated good performance in predicting inundated depth under the observed and scenario rainfall events.The spatial inundated depths predicted by the MORF model were close to those of the coupled model,with differences typically less than 0.1 m and an average correlation coefficient reaching 0.951.The MORF model,however,achieved a computational speed of 200 times faster than the coupled model.The overall prediction performance of the MORF model was also better than that of the k-nearest neighbor model.Our research provides a novel approach to rapid urban flood mapping and flood early warning.

Zero increase in peak discharge for sustainable development

For urban land development, some or all natural land uses （primarily pervious） are converted into impervious areas which lead to increases of runoff volume and peak discharge. Most of the developed countries require a zero increase in peak discharge for any land development, and the policy has been implemented for several decades. The policy of zero increase in peak discharge can be considered as historical and early stage for the low impact development （LID） and sustainable development, which is to maintain natural hydrological conditions by storing a part or all of additional runoff due to the development on site. The paper will discuss the policy, the policy implementation for individual projects and their impact on regional hydrology. The design rainfalls for sizing LID facilities that are determined in 206 weather stations in USA are smaller than design rainfalls for sizing detention basins.The zero-increase policy links to financial responsibility and sustainability for construction of urban stormwater infrastructures and for reducing urban flooding. The policy was compared with current practices of urban development in China to shine the light for solving urban stormwater problems. The connections and differences among LID practices, the zero-increase policy, and the flood control infrastructure were discussed. We promote and advocate the zero-increase policy on peak discharge for comprehensive stormwater management in China in addition to LID.