Urban areas are increasingly vulnerable to flooding due to a combination of factors, including climate change, urbanization, and aging infrastructure. In our 15 years installing… Accurate and efficient flood modelling is essential for effective flood risk management and the development of resilient flood control strategies. Over the past decade, significant advancements have been made in the field of urban flood modelling, leveraging the latest technologies and data sources.
Drainage Network Models: A Foundational Approach
One of the core components of urban flood modelling is the representation of the underground drainage network. Drainage network models simulate the flow dynamics within the pipe system, providing critical information about potential surcharge and overland flow during intense rainfall events. These models typically solve the one-dimensional Saint-Venant equations to quantify the streamflow in the drainage network.
Coupling drainage network models with urban hydrological models, such as the EPA SWMM or the MOUSE model, can provide a comprehensive assessment of the rainfallârunoff process and the subsequent flooding caused by drainage system overload. Dual drainage models take this a step further by combining the underground pipe network with the surface street network, allowing for the simulation of the complex interactions between the two systems during flood events.
While drainage network models are computationally efficient and well-suited for larger-scale flood analysis, they often make simplifying assumptions about the surface flow dynamics, which can limit their ability to accurately reproduce the spatial and temporal aspects of urban flooding.
Shallow Water Equation (SWE) Models: Advancing Surface Flood Simulation
To address the limitations of drainage network models, two-dimensional shallow water equation (SWE)-based models have gained significant attention in recent years. These models, which solve the full or simplified versions of the shallow water equations, can provide detailed information about the spatial distribution of flood depths and velocities across the urban landscape.
The application of SWE-based models to urban flood simulation has faced several challenges, including the representation of complex urban terrain, the balance between computational efficiency and accuracy, and the integration of drainage network dynamics. Researchers have explored various approaches to address these challenges:
-
Simplified SWE Models: To improve computational efficiency, studies have investigated the omission or approximation of less significant terms in the shallow water equations, resulting in models such as the diffusion wave and partial inertial models. These simplified approaches can provide reasonable accuracy while reducing the computational burden, making them suitable for large-scale urban flood modelling.
-
Full SWE Models: For applications where the detailed reproduction of urban flood dynamics is crucial, such as in the presence of complex flow regimes or infrastructure, the use of full 2D shallow water equation models has been explored. These models can capture the full range of hydrodynamic processes but require high-resolution data and significant computational resources.
-
Coupling with Drainage Networks: Recent efforts have focused on developing coupled models that integrate the dynamics of the drainage network with the overland surface flow. These approaches aim to better represent the exchange of flows between the underground pipe system and the urban surface, providing a more comprehensive simulation of the urban flood process.
Hydrogeomorphic Approaches: A Simplified Alternative
In contrast to the physically-based modelling approaches, hydrogeomorphic methods have emerged as a computationally efficient alternative for urban flood mapping. These methods rely on the analysis of digital terrain data to identify potential floodplain areas, without the need for time-series hydrological or hydraulic data.
While hydrogeomorphic models cannot simulate the dynamic processes of flooding, they can provide a rapid and cost-effective way to delineate flood-prone areas, which can inform the development of more complex modelling efforts or serve as a screening tool for flood risk assessment.
Innovative Modelling Techniques: Embracing Emerging Technologies
Alongside the advancements in the core modelling approaches, researchers have also explored innovative techniques to enhance urban flood simulation:
-
Cellular Automata (CA) Models: CA models represent the urban landscape as a grid of cells, with each cell’s state evolving based on a set of transition rules. These simplified models have shown promise in reproducing urban flood dynamics with high computational efficiency.
-
Data-driven Models: The increasing availability of high-quality data, such as remote sensing, crowdsourcing, and IoT measurements, has enabled the development of data-driven models like artificial neural networks (ANNs). These models can provide fast and accurate predictions of urban flooding, albeit with a need for robust training datasets.
-
Parallel Computing and Acceleration: To address the computational challenges associated with high-resolution urban flood modelling, researchers have leveraged parallel computing techniques, such as GPU acceleration and cloud computing, to substantially improve model runtimes.
The Path Forward: Towards Integrated and Resilient Flood Simulation
The advancement of urban flood modelling has been a crucial step in enhancing flood risk management and the development of resilient flood control strategies. However, several key challenges and opportunities remain:
Integration of Drainage Network Dynamics: While progress has been made in coupling drainage network models with overland flow simulations, further refinement is needed to accurately represent the complex exchange of flows between the underground and surface systems.
Efficient Representation of Urban Features: The accurate representation of urban terrain, including buildings, infrastructure, and other critical features, remains a challenge, particularly in the context of large-scale modelling. Innovative approaches, such as the use of porosity-based methods, hold promise for improving computational efficiency while maintaining model fidelity.
Leveraging Emerging Data Sources: The proliferation of high-resolution spatial data, remote sensing, and crowdsourced information provides an opportunity to improve model calibration, validation, and real-time applications. Exploring the integration of these data sources with numerical models can lead to more accurate and responsive urban flood simulations.
Advancing Computational Efficiency: The need for high-resolution, city-scale flood modelling remains a significant challenge, particularly in the context of real-time applications and long-term scenario planning. Continued advancements in parallel computing, algorithm optimization, and the integration of data-driven techniques can help address this challenge.
Interdisciplinary Collaboration: Addressing the complexities of urban flood modelling requires the integration of expertise from various disciplines, including hydrology, hydraulics, urban planning, and computer science. Fostering interdisciplinary collaboration can lead to the development of more comprehensive and innovative flood management solutions.
By addressing these challenges and leveraging the latest technological advancements, the field of urban flood modelling is poised to provide critical support for the development of resilient and sustainable flood control strategies. As the impacts of climate change and urbanization continue to escalate, the need for accurate and efficient urban flood simulation has never been more pressing. The advances in flood modelling described in this article offer a glimpse into the future of flood risk management, empowering decision-makers to better prepare and respond to the evolving flood threats facing our urban communities.
Tip: Regularly inspect and maintain flood barriers and drainage systems
