Predicting long-term flooding trends with AI


28-10-2019
  Bernard Taylor, Chairman of the Royal Commission for the Exhibition of 1851, presents PhD student Robert Rouse with his Industrial Fellowship  Credit: Royal Commission for the Exhibition of 1851

University of Cambridge PhD student Robert Rouse has been awarded an Industrial Fellowship to develop machine learning capable of predicting future flooding events.

Robert is a student with the EPSRC Centre for Doctoral Training (CDT) in Future Infrastructure and Built Environment, working in Artificial Intelligence (AI) for Environmental Risk. He has been named as one of the UK’s most promising young doctoral engineers by the Royal Commission for the Exhibition of 1851, securing up to £80,000 to enable him to develop machine learning-based models that will be capable of predicting flooding years from now, and in greater detail.

Robert will investigate where AI can provide a detailed understanding of how variables such as rising temperatures and rainfall will affect water levels far into the future. His industrial partner, Mott McDonald, a global engineering, management and development consultancy, can then use the data this model will produce to advise local authorities on how to work with the elements within their control, namely infrastructure and water management, to help build resilience against climate change and its effects. Together they aim to produce a system that is as user-friendly as possible and which incorporates user requirements from the outset.

The project, funded for up to three years, will ideally lead to a patent or substantial business development, allowing Robert to conduct impactful research while gaining valuable industrial experience. 

“Climate change is bringing us wetter weather, potentially more extreme weather, and, as a result, increased flood risk, which is why we need to understand the extent of that risk to help us prepare the surrounding infrastructure and environment,” said Robert. “There is a pressing need to understand future trends, particularly for data-poor regions, and it is these regions that are usually most at risk from the adverse effects of climate change. Detailed predictions of impact and risk, however, would enable local authorities to prepare effective and lasting solutions for their communities, and investigating whether or not machine learning models can offer an advantage here, is key to my project."

“The Industrial Fellowship programme will bring my research closer to application, enabling me to work faster and more flexibly with greater computational and information resources, given the huge amount of data involved. I also hope to benefit from the programme’s network of inter-disciplinary experts," he added.

Bernard Taylor, Chairman of the Royal Commission for the Exhibition of 1851, said: “Supporting British innovation is more important than ever before. These Industrial Fellowships provide an opportunity for British businesses to conduct cutting-edge research and development, and develop intellectual property at reduced cost. This year’s Industrial Fellows include some of the brightest minds in their fields and they represent the huge diversity of talent in science and engineering that Britain has to offer.”

Applications for the 2020 Industrial Fellowships are now open.

Image: Bernard Taylor, Chairman of the Royal Commission for the Exhibition of 1851, presents PhD student Robert Rouse with his Industrial Fellowship

Credit: Royal Commission for the Exhibition of 1851

Reproduced courtesy of the University of Cambridge Department of Engineering

To read more information, click here.

The University of Cambridge is acknowledged as one of the world's leading higher education and research institutions. The University was instrumental in the formation of the Cambridge Network and its Vice- Chancellor, Professor Stephen Toope, is also the President of the Cambridge Network.

University of Cambridge (cam.ac.uk)