Department of Engineering / News / Diving into AI: exploring the potential for AI to enhance water quality management 

Department of Engineering

Diving into AI: exploring the potential for AI to enhance water quality management 

Diving into AI: exploring the potential for AI to enhance water quality management 

Group of people in kayaks among reeds on the autumn river.

Researchers from the University of Cambridge convened a workshop and produced a policy brief to identify opportunities to leverage AI to improve water quality.

It is essential that academia, government and the industry work together to leverage new opportunities that AI offers. This workshop is the first step in developing a community of practice on these topics and an improved identification of the knowledge gaps that are most relevant to water policy in the country.

Dr Edoardo Borgomeo

Managing the threat of water pollution is central to public health and environmental quality in the UK. While significant improvements in water quality have been attained, water quality across the country still does not meet regulatory standards.

“Climate change, aging and inadequate infrastructure, increasing concentrations of forever chemicals and pathogens mean that the nation’s water quality is under pressure. Moreover, as communities’ interests and expectations for better water quality increase, so does the need for decision-makers to drive significant improvements in the nation’s rivers and seas” said Dr Edoardo Borgomeo from Cambridge’s Centre for Sustainable Development. “We partnered with the Environment Agency to organise a one-day workshop focused on understanding if and how AI and machine learning can be leveraged to address these issues.”

As the environmental regulator, the Environment Agency (EA) leads efforts to monitor water quality across the country and regulate practices that might lead to increased pollution. Luke Holmes Senior Research Scientist in the Environment Agency’s Chief Scientist’s Group, said: “The workshop was very helpful in beginning to develop a framework for assessing the potential of AI to contribute to our mission. It was also a great opportunity to connect with researchers and practitioners interested in advancing the responsible use of AI in environmental sciences’.

The workshop was supported by an EPSRC Impact Accelerator Award.

The policy brief: 'Diving into AI? Exploring the potential for AI to help deliver clean rivers, lakes and seas in England' generated as a result of the workshop is available at the following link: https://doi.org/10.17863/CAM.118743

Participants to the workshop included representatives from academia, the public sector and industry. The workshop identified four core priority areas where AI might contribute to managing water quality in the UK:

  • Operational efficiency. Managers and regulators are tasked with making cost-effective and timely decisions with regards to water quality, such as issuing water quality forecasts, prioritising inspections or collecting data on new, emerging contaminants. The workshop explored ways in which AI can support these decisions, for example by helping optimize the location of sensors or by harmonizing datasets with different spatial and temporal resolutions.
  • Modelling and understanding. To inform water quality management, we also need to develop novel understanding of the status and trends of water quality. This entails, for example, understanding how water quality might change under alternative climate change and land-use scenarios, or modelling which factors are more significant in influencing sewer overflows in different locations. Under certain conditions, AI offers opportunities to improve our understanding of water quality process, thus enabling for better decisions and predictions in the future.  
  • Institutional knowledge and trustworthiness. Water is a public resource and therefore all decisions made with respect to water should be transparent and understandable. Can AI help improve the transparency and trustworthiness of decisions with respect to water quality? The workshop explored the limits and potential for AI approaches to make environmental models more understandable to users, and to also improve our understanding of the reliability of predictions.
  • Last mile and dissemination. Citizens and the natural environment are the ultimate beneficiaries of any decision made with regards to water quality. The workshop identified ways that AI can help provide better and more timely information to the public, and also improve citizen engagement through, for example, citizen science and data collection. 

“It is essential that academia, government and the industry work together to leverage new opportunities that AI offers” said Dr Borgomeo “This workshop is the first step in developing a community of practice on these topics and an improved identification of the knowledge gaps that are most relevant to water policy in the country.”

The text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified.  All rights reserved. We make our image and video content available in a number of ways that permit your use and sharing of our content under their respective Terms.