Predicting Flood Risks to City Infrastructure Systems Utilizing Scalable, Time Sensitive Modeling

Abstract

Flooding is emerging as the most expensive and frequent natural hazard around the world. Floods are highly dynamic in nature and cause physical damage to our built environment, loss of life, economic damage, and major impacts to society. An example of this is the at-ground road system, which comprises 30-60% of a city’s area in the US, is highly susceptible to flood damage, while still needs to act as evacuation routes for local residents. Similarly, the underground built system is extremely vulnerable to flooding damage as well as life risk to anyone within it. With urban landscapes constantly evolving, accurately predicting flood propagation and extent is imperative to mitigate these risks, especially as floods worsen due to climate change.

Historically, the focus of flood risk assessment through industry and academia has been on the coastal urban environment, assessing the impact of fluvial flooding. This resulted in many risk assessment tools that mostly caters to the infinite amount of flood water identified from a riverine or coastal fluvial flooding. As for the rain-driven impact, the common practice simply changed the flood modeling to pluvial oriented, keeping the rest of the risk tool components identical for the different flood mechanisms. For pluvial flooding, existing urban flood modeling tools such as SWMM and PC-SWMM are limited by their catchment-based approach, neglecting surface runoff dynamics and spatial-temporal flood impacts. Consequently, these tools fail to capture the full extent of rain-driven floods, underestimating their severity and impact on urban environments.

Addressing this gap requires sophisticated simulations that account for rain event characteristics and city morphology, yet such simulations are computationally demanding and require detailed urban data. Currently, flood impact analysis tools lack specificity for pluvial flood risks and do not address the risks to various city systems beyond building damage. As a result, the contribution of pluvial floods to overall flood risks is underestimated, compromising infrastructure resilience. As flood model results are a critical component in flood risk assessments, the accuracy of spatial temporal urban flood results will allow the pluvial flood impact assessment to be simplified and the flood damage to the different urban systems will be quantified.

This research aims to develop a scalable and streamlined method to accurately quantify the risks of rain-driven floods to urban infrastructure systems. It addresses three key questions: (1) To what extent does current practice underestimate pluvial flood impacts? (2) What are the impacts of pluvial flooding on pavement systems when incorporating spatial-temporal modeling? (3) What is the significance of modeling pluvial floods using urban underground spaces? Using advanced flood modeling and numerical soil-water infiltration techniques, this research will quantify damages and lifecycle impacts to pavement and underground spaces systems. The method will provide information on the spatial and temporal distribution of flood damage and will enable scaling up single-element assessments to system-wide impacts. This holistic approach will improve urban flood risk management, supporting informed decision-making and the development of resilient infrastructure systems.