Selection of urban flood shelter locations based on risk assessment

Abstract

Within the framework of global climate change, there is a recurring occurrence of floods and waterlogging disasters, which pose significant risks to human lives and overall safety. The phenomenon of urbanization has led to an increased vulnerability of cities to flood disasters. Evacuation serves as a viable strategy for adapting to climate change, managing intense precipitation events, and mitigating the impact of flood disasters. Hence, the meticulous choice of shelters has emerged as a crucial element in enhancing the effectiveness of evacuations. Nevertheless, the process of choosing shelters sometimes neglects to consider the potential for flooding and has not undergone thorough testing. Hence, the objective of this research is to systematically choose shelters within the central metropolitan region, incorporate flood risk considerations into the shelter site model, and assist cities in mitigating the consequences of floods resulting from climate change. To begin with, employ a two-dimensional shallow water model in order to replicate instances of intense precipitation and produce data on flood inundation. Subsequently, the simulation results are utilized to assess the flood risk using the entropy weight approach. This risk is then incorporated into the multi-objective optimization model to determine the optimal shelter placement. Ultimately, the ABM model was employed to create a simulation of crowd evacuation in urban areas, confirming the efficacy of shelter selection and thoroughly examining the influence of various shelters on the efficiency of community emergency evacuation. The study's findings indicate that a total of 26 shelters were chosen within the study area using a multi-objective methodology, ensuring the inclusion of all affected populations. The utilization of multi-objective models in the selection of shelter places has been shown to enhance evacuation efficiency, as confirmed by the ABM model. The simulation findings indicate a decline in the population's death rate from 1.94% to 0.98%, an increase in the evacuation rate from 97.4% to 98.4%, and a reduction in the number of casualties by 133 individuals.