Skip to the content

Ordnance Survey tests AI tool to measure water pollution


Mark Say Managing Editor

Get UKAuthority News


River with surface pollution
Image source: Morris

Ordnance Survey (OS) and CGI have developed an initiative using a new AI model to remotely detect from space sewage overspills.

The national mapping and IT and business consultancy are working on the project under the latter’s Sustainability Exploration Environmental Data Science (SEEDS) programme.

The first phase has involved the creation of an AI and satellite tool, following which the project will provide a proactive approach to environmental management and nature protection in the North Devon UNESCO Biosphere reserve, which is centred on Braunton Burrows, the largest sand dune system in England.

It will involve bringing together multiple sources of data – including from sensors, the UK Environment Agency, satellites and the project partners – to identify where water pollution takes place or may do so in the future.

The AI algorithms will be used to map out where incidents of pollution have taken place over time using CGI’s GeoDat360 platform.

Proactive environmental management

Mattie Yeta, CGI’s chief sustainability officer, said: “Following a successful first phase of the project, which led to the creation of an AI and satellite tool that can predict pollution events with up to 91.5% accuracy, we are excited to launch this second phase, which provides an innovative and proactive approach to environmental management and nature protection.”

She added: “The solution will benefit farmers, governments, water companies and other stakeholders by protecting our water from pollution and contamination, which is vital for both our way of life and the life of our waterways and coastlines.”

Donna Lyndsay, strategic market lead, environment and sustainability at OS, commented: “With the innovative technical ability that both CGI and OS have in abundance, we believe we may be able to help provide actionable insights for addressing water pollution events.

“We are looking forward to testing the AI models produced in our previous phase to see how they perform in the real world. We want to see if our work can be really used to support the biosphere needs in identifying interventions to help preserve and restore our valuable habitats.

“If successful this has the potential to support remote monitoring of UK waterways for signs of pollution using data that is objective, regularly updated, and scalable.”

A long term element of the SEEDS programme plan is for the tool to be replicated and rolled out to reduce the impact of raw sewage pollution globally.

Register For Alerts

Keep informed - Get the latest news about the use of technology, digital & data for the public good in your inbox from UKAuthority.