Dynamic disaster risk assessment of urban waterlogging on pedestrian flow by intelligent simulation of hydrodynamics coupled with agent-based models in Chao-yang river basin of Nanning, China

Global climate change leads to an increase in the frequency and intensity of extreme rainstorms. At present, in China, which is experiencing rapid urbanization, urban flooding triggered by heavy rainstorms has emerged as a prominent issue, exerting far-reaching influences on socio-economic development, ecology, and people’s livelihoods. In response to this, China has put forward the concept of sponge cities and has shifted from pilot projects to comprehensive and systematic citywide implementation, with the aim of strengthening urban resilience in rainstorm management. This article takes the Chaoyang River area in Nanning City, South China as an example and proposes a dynamic risk assessment framework integrating hydrodynamic models and ABM to reflect flooding and pedestrian response to rainstorms. The research results show that under the design rainstorm scenarios with return periods of 5 years and 30 years, the rainstorm-induced flooding process in the study area presents a dynamic evolution pattern. It develops rapidly to the extreme or severe hazard grade within 1–2 h, and then declines slowly and persists until 8 h. The exposure and sensitivity of pedestrian mobility to flooding disasters extend across most areas of the study area within 1–5 h during the rainstorm-induced flooding process, with medium, high, or extreme risk levels observed during the 2–3 h period.Among the affected sensitive pedestrians, the gender ratio is roughly equal, and the proportion of the elderly and child populations is as high as 46.5%. The overall disaster resilience capacity of the study area is significantly insufficient, leading to a predominance of medium, high, or extreme risk levels for pedestrian mobility during the 2–3 h period. In conclusion, the study area should accelerate the construction of various sponge facilities on the underlying surface and comprehensively enhance diverse emergency management measures for excessive flooding disasters. The dynamic disaster risk simulation and assessment techniques proposed in this article can serve as essential scientific support for the construction of a digital twin system in China’s sponge cities, reflecting both virtual and real scenarios and facilitating comprehensive resilience capabilities such as forecasting, warning, simulation, and contingency planning for intelligent rainstorm management.