A Numerical Approach for the Identification of Flood Vulnerability of a Metro System in a Mega-City

With the urbanization of mega-cities, metro lines have been successfully explored as a viable option for the improvement of mobility. However, depending on the climate regime this type of infrastructure, may be exposed to severe flooding from extreme events. Having in mind the documented increase in both intensity an frequency of floods, it is therefore necessary for mega-cities to identify flood vulnerabilities of this mass transportation system. Moreover, as all around the globe flood disasters have swept billions of dollars of damage, resulting from the inundation of underground metro infrastructures (e.g., metro tunnels and facilities) (i.e. Guangzhou 2016, Manila 2017, New York 2014, Mexico 2017) .

In recent years, Mexico city has suffered from catastrophic flood disasters resulting from convective storms. For instance, on June 14, 2017, an extreme storm (60mm in 1hr) occurred in one of the main lines of the metro system affecting several metro stations. This even caused catastrophic damage to metro equipment, and resulted in the suspension of the service for several hours. The level of damage in this part of the city has forced local authorities to evaluate flood vulnerability of its metro system to heavy rainfall events. As one of the most hazard-prone and densely populated cities in the world, Mexico City faces significant risk of loss of lives and challenges to development progress as a result of natural disasters.

This study will present the use of an open-source model named Itzï that enables the 2-D numerical modelling of urban floods integrated with the open-source geographic information system (GIS) software known as GRASS. Results from this tool will be used as an input for another open-source tool that produces realistic natural hazard impact scenarios to enable better planning, preparedness and response activities. This tool entitled Inasafe, provides a simple but rigorous framework to utilise scientific data to improve decision making processes of local authorities. Realistic hazard scenarios based on a recorded events for which high quality data is available, will be employed for model calibration and validation. It will be shown that results from this methodology enable a better disaster management planning of the city to extreme precipitation events.