Extreme rainfall induced risk mapping for metro transit systems: Shanghai metro network as a case

Abstract

In recent years, a changing climate has induced flood risk as a great threat to the safety and reliability of the metro transit network in mega-cities. A highly networked metro system can lead to a quick spread of this risk, and furthermore, the impact range of single-node accidents of a network is nonlinearly amplified through network connectedness defined by its topology. This study proposes a risk assessment framework integrating extreme rainfall simulation and network loss analysis.​​ The methodology employs the Areal Reduction Factor (ARF) and Soil Conservation Service Curve Number (SCS-CN) to model rainfall-induced flooding, ​​coupled with a multi-layer network-based approach that distinguishes topological interactions between stations and lines.​​ Taking Shanghai metro as an example, this paper highlights its risk follows an exponential distribution to extreme rainfall events, characterized by the finding that nearly 50 % of extreme rainfall events result in <5 % network loss, whereas fewer than 5 % of the events lead to >50 % network loss. When rainfall centers are located in the urban center where metro stations are densely distributed and intricately connected, or when the rainfall intensity and the spatial distribution uncertainty increases, it will pose a greater risk to the metro network.