Modelling under water

Flooding is a highly complex phenomenon which needs to be considered at each stage of the planning, design and building process, Justin Butler from Ambiental gives more details about his company’s work in this area.

UK Based flood modelling and flood risk consultancy Ambiental has grown since its inception 12 years ago and has offices in Australia and Singapore. Numerous multinational organisations, engineers, construction companies, insurers and government bodies now rely on Ambiental data, reports and modelling to improve their flood planning, designs and decision-making.

Ambiental aims to reduce the time/cost associated with flood modelling for dam engineers and consultants, as well as to provide a more accurate and definable approach to quantitative risk assessment (QRA). It uses simulation technologies to better analyse and quantify future … And the impact of different infrastructure/dam breach scenarios.

For insurers and finance companies, the company’s data, reports and visualisation tools provide increased confidence, which helps them to make better decisions regarding insurance premiums and potential losses. This, in turn, helps to better manage risk and returns value back to the business.

With flooding regarded as one of the most significant natural catastrophic perils faced by businesses and communities alike, especially given recent events worldwide, Ambiental’s models and data are increasingly sought after.

Indeed nearly half (45%) of all weather-related events in the last 30 years or so have been floods, followed by storms (30%) and forest fires, heat waves and droughts (16%). A total of 120,000 people have lost their lives since 1980 as a result of these weather catastrophes, 57% from flooding alone and 39% from storm events. |For of the five costliest weather-related catastrophes are also attributed to floods. At US[illegible]., floods accounted for 56% of overall losses, followed by storms with overall losses of US[illegible] (…%) (Source: MunichRE).

As Ambiental’s Founder and CEO Justin Butler commended: “Flooding is a highly complex phenomenon which needs to be considered at each stage of the planning design and build process. With recent improvements in computing power, data availability and distribution platforms, this means that a wider range of specialists and non-specialists alike can access out flood models, maps and reports more quickly and cost effectively than before. This can also help our clients in the building process. We routinely product high impact visuals to support project tenders as well as more detailed analytical work.”

Given that flood analysis is only one part of an infrastructure or risk assessment project, many of Ambiental’s clients find that it doesn’t make business sense to build and sustain in-house teams of flood experts. Ambiental provides a ‘just in time’ service to most their clients’ needs in a more flexible way than they have often been able to, says the company.

Strong focus

With a strong focus on R&D Ambiental’s goal is to map entire countries and continents, as well as individual areas of interest with, it says, unprecedented detail and accuracy through a combination of powerful modelling software and high quality input data. Ambiental’s links to engineering consultants, insurers, academia and regulatory policymakers have helped to secure its position in the flood risk industry.

Ambiental innovations include the recent launch of Realistic Event Modeller – and innovative, practical and relevant solution to the challenges of transparency faced by all major global financial institutions today.

Another innovation is Flowroute, a proprietary flood modelling platform which consists of a core engine and digital flood map ‘production-line’ Which enables the company to create high-detail three-dimension flood risk maps in a highly efficient manner, almost anywhere on earth where data is available.

Originally conceived some ten years ago in conjunction with Cambridge University, the software can be used to evaluate flood probabilities, depths and velocities.

Flowroute has the capability to combine precision with flexibility in two-dimensional ‘+’ flood modelling. It is designed to be able to handle complex environments and a variety of flooding types including river, coastal, and pluvial (surface water) sources. The software is also particular effective for simulating multiple dam burst scenarios accurately and cost effectively.

Flowroute incorporates high detail topographic information at the sub-meter level which includes buildings, structures and flood defences as well as the impact that these have upon patterns of flood risk for large urban areas. It is ideal for high resolution LiDAR based modelling and is easy to use for practitioners that don’t have a background in flood modelling. By working seamlessly with all major GIS platforms, Flowroute can be used to extract additional value from other data assets and mapping collected as part of an engineers’ everyday project work.

Simulated water flows for multiple probabilities and return periods are routed along streets and around buildings. The model provides information including depth, duration, velocity and extent of flood risk down to the individual building level. This is particularly important for urban planning, risk mapping, insurance underwriting and loss/damage assessment. Similarly because of the three-dimensional nature of the underlying topographic information, new buildings and proposed future developments can be incorporated into the model to enable assessment of future patterns of flood risk, including drainage and flood defences.

Flowroute is based on well-established hydrological and hydraulic principles which can be adapted to incorporate future climate change scenarios such as, increasing sea levels or more storm events. Ambiental regularly update river and floodplain dynamics with complex urban and rural environments, as well as in the mountainous regions of the world where other broad-scale modelling approaches often experience computational stability problems. Model outputs are analysed and communicated more effectively using a variety of visualisation tools and options.

A core study of the Brisbane flooding in 2011 demonstrated Flowroute’s capabilities in terms of predictive accuracy. With estimated damages of US$2.3B and an impact on Australias GDP of US$30B. the Brisbane flooding represented a major event, within the 1-in-100 year profile. At the time of the flood most insurers, planners, and flood evaluators reemployed a commonly used dataset which, as our tests demonstrate, is nowhere near as accurate and comprehensive as it needed to be. Ambiental’s 2011 model of the Brisbane flooding in Australia achieved 95% accuracy compared with the existing flood model at 37%. Almost two in three flooded properties had previously been missed by the existing flood model compared with just 1 in 20 being missed by the Ambiental model.

Ambiental is able to produce highly accurate data by isolating flooding risks from differing sources. Intense rainfall-driven urban “pluvial” or surface water flood events, water from point source inflows (e.g. dam or defence breech/overtopping, surcharging sewers, pipe bursts etc) to achieve the most realistically possible enhanced simulation models.

High resolution flooding analysis has become an important part of the dam industry. In the UK, for example, the Environmental Agency developed a similar approach as part of a nationwide Reservoir Foundation Mapping (RFM) study. Around 2000 reservoirs over a given threshold size were modelled to produce maps of potential inundation extents and depths due to dam burst. This information was then made publicly available to assist planners and emergency responders and could not have been undertaken on such an ambitious scale even ten years ago due to computational limitations.

Although reservoir flooding is unlikely, each year there are a small number of incidences (14 recorded between 2004 and 2008 in the UK) that are seemed serious enough to require emergence dressdown of the reservoir emergency works and/or evacuation of people in the potential path of flooding. One of the most high profile of these incidents was in the summer of 2007 at the Ulley reservoir in Rotherham, UK.  In this case extreme rainfall caused damage to the dam and the risk of reservoir flooding let to a decision to evacuate around 1000 people near the reservoir and to close the roads. This highlighted the risk to communities being in reservoir flood areas and prompted a government review in 2007, which recommended that the government takes measures to improve reservoir emergency preparedness. The reservoir flood maps produced are ‘outline’ maps meaning that they show only the areas that might be flooded in a realistic worst case reservoir failure scenario.

These maps do not give any indication about the likelihood of flooding depth or speed of the flood waters or the length of time it would take for the waters to reach any location – due to the sensitive nature of the information from a national security perspective, this information is not publicly available.

Flowroute has been continually optimised to reduce processing times and leverage modern computing capabilities. Indeed, Ambiental was the first company in the world to carry out flood modelling in the Cloud, suing the Amazon Elastic Compute (EC) 2. The software can be run in a distrusted or cloud computing environment. It is capable of simulated flood flows across hundreds of millions of sdl domains in parallel using standard desktop computers, providing optimum flexibility across small or large projects.

Today, Ambiental’s services including Flowroute are employed worldwide by major insurers, governments, property planners, local government and commercial organisations, all of whom share a common need to better understand flood risk. With the recent flooding in places such as the UK, there is an increasing need for large-scale infrastructure and emergency response planning and Ambiental data solutions are increasingly being deployed in-house and ‘on-demand’ to support decision making this regard.