Risk of flooding from underground water and waste water infrastructure (CADDIES)

KTP Project with ICS Consulting

Water company costs for flooding caused by surcharging sewers and burst water mains can be significant when including claims, insurance costs and outcome delivery incentives penalties. For example, one burst 20 inch water main in London’s Tooley Street in 2008 alone caused losses of “tens of millions” of pounds for Thames Water (Evening Standard, 2013). These events can also be devastating to customers as they pose substantial social and economic effects that may continue over extended periods of time. Understanding the potential risk of flooding from buried assets is essential in estimating risk exposure. However, the provision of a timely, accurate and comprehensive assessment of flood risk is challenging as it requires complex modelling and analysis methodologies. ICT solutions are required that allow risk of flooding from pipes to be assessed on a network wide scale. Following on from the development of a generic two-dimensional (2D) flood modelling tool (CADDIES) through funding from EPSRC, the researchers from the University of Exeter’s Centre for Water Systems have teamed up with ICS Consulting to customise and integrate the tool within ICS own Asset Data Management System (ADMS) to enable fast and accurate modelling and assessment of flood risk.

The two existing best practice approaches involve either one-dimensional (1D) or 2D modelling. However, with either there is an unacceptable trade-off between the speed and accuracy when multiple runs are required. The 1D models are fast but suffer from oversimplification of flood flows which are assumed to be unidirectional. However, actual flooding results in divergent/recombining paths and ponding, which limits the 1D model’s capability to accurately and realistically predict flood depths, velocities and extents, all key indicators of risk to life and property.  While being more accurate, the existing 2D models are computationally expensive, requiring hours or even days to complete each simulation.

Using the new concept of “Cellular Automata”, University of Exeter researchers developed a fully dynamic 2D model (CADDIES).  This takes advantage of the local interaction between water levels surrounding each square grid cell, the huge amount of freely available high-resolution LiDAR data and the power of Graphical Processing Units (GPU). As a result, the ICS-implemented system produces complex dynamically 2D flood extents  and ponding areas at the very fastest speeds, while maintaining the highest accuracy levels. It closely matches the output of industry standard commercial software and is between 5-20 times faster than conventional 2D methods.

There are multiple benefits from the developed CADDIES methodology as implemented within ICS asset planning and management suite of tools, ADMS:

  • Better understanding of sources and consequences of flood risk from their own assets will allow water service companies (WSC) to prioritise their investment in flood risk mitigation making better use of limited financial resources;
  • Greatly enhanced run times of up to 20 times better than that available with existing commercial 2D models allows use in real-time company operations responding to actual flooding events;
  • Consistent network wide risk analysis results produced by ADMS, using a single setup and tens of thousands of analyses in one batch run, as opposed to running each analysis individually and by a number of analysts;
  • Reduced harm to local economies from flood events inundating residential and commercial properties, and interconnected multi-utility infrastructure through targeted investment;
  • Scientifically tested and proven flood modelling with automated time step optimisation to achieve better accuracy than 1D models and much faster runs than 2D models;
  • The use of structured square grid, which is readily available from LiDAR data and automatically used by CADDIES, reducing time-consuming processing associated with unstructured grids and avoiding a large amount of human intervention.

Water distribution and wastewater systems in the UK consist of over 700,000 km of water distribution and sewer pipes, which represents a large risk exposure from flooding caused by sewer surcharging or water main failures. There is also an emerging abundance of freely available high resolution (one meter or less) LiDAR data due to the advent of remote sensing, which enables wider applications of detailed flood risk modelling and analysis. Considering the above asset base and the need to better assess the consequence of flooding in urban areas, the potential for improving the risk assessment processes and reduce harmful consequences of flooding is enormous.

As an example, ICS have already performed flood risk analysis for a major WSC, over 20,000 water pipe simulated failure locations were considered. Further work on 120,000 pipe failure locations for the same WSC in under way.  

ICS offer the tool as a fully integrated software package, either for client use or as a service, to allow wider adoption of the methodology and the associated tools.

Until recently flooding from sewer and water distribution networks has mainly been considered reactively, after the flooding has occurred. Flooding from water mains or sewers is often an unpleasant and distressing event for customers, with potential to waste precious water resources, cause pollution and harm people and the environment. Infrastructure and businesses can also be affected by flooding, further escalating the impact on local communities.

The ADMS/CADDIES methodology helps mitigate the threat of flooding to people, their property and infrastructure by providing fast and accurate data for decision makers. These significant enhancements over current technologies provide accurate information to water companies allowing them to make better investment and operational decisions.  By improving the intelligence about flooding risks both quickly and accurately, the result will be better management of natural water resources and an enhanced ability to mitigate flooding impacts on society and the environment. At a time of increased pressures from severe flooding due to factors like climate change these new technologies provide water companies with leading edge tools for understanding and managing flood risks and impacts.

Modern high-performance computing used by ADMS/CADDIES minimises energy usage and increases carbon efficiency of the hardware used to run extensive flood risk analyses.

Back to Urban drainage system management