Using AI to avert ‘environmental catastrophe’

Funded by UK Research and Innovation (UKRI), the Centre for Doctoral Training in Application of Artificial Intelligence to the study of Environmental Risks (AI4ER) was one of 16 new Centres for Doctoral Training (CDTs) announced earlier this year. The Cambridge Centre will be led by Professor Simon Redfern, Head of the Department of Earth Sciences.

Climate risk, environmental change and environmental hazards pose some of the most significant threats we face in the 21 century. At the same time, we have increasingly larger datasets available to observe the planet, from the atomic scale all the way through to global satellite observations.

“These datasets represent a transformation in the way we can study and understand the Earth and environment, as we assess and find solutions to environmental risk,” said Redfern. “Such huge datasets pose their own challenges, however, and new methods need to be developed to tap their potential and to use this information to guide our path away from environmental catastrophe.”

The new Centre brings computer scientists, mathematicians and engineers together with environmental and geoscientists to train the next generation of thought leaders in environmental data science. They will be equipped to apply AI to ever-increasing environmental data and understand and address the risks we face. The first cohort of PhD students will start their studies in October..

At the same time as human-induced climate change becomes increasingly apparent, urbanisation and the growth of megacities generate other risks, as society becomes potentially more fragile and vulnerable to geohazards such as earthquakes, volcanic eruptions, floods and tsunamis. Alongside satellite data, autonomous sensors, drones, and networks of instruments provide increasingly detailed information about such risks and their potential impacts.

Examples of the projects already underway that apply AI methods to exploring environmental risk include the use of satellite observations to chart the distribution and pathways of whales through the oceans^1, large datasets to understand biodiversity changes in woodland habitats, machine learning to understand earthquake risk² and the use of drones to monitor hazards at active volcanos.³

Cambridge is already a world leader in artificial intelligence and machine learning research⁴, and many of its AI researchers work alongside world leaders in environmental monitoring and modelling, including from the British Antarctic Survey and elsewhere at the University.

The new centre combines this work with the interests of dozens of external partners including Microsoft, DeepMind, The European Development Bank, Friends of the Earth, the European Space Agency, the Environment Agency, resource industry leaders and policy partners, to form an outstanding alliance focused on leading the next generation of environmental data science forward.

The new Centre is part of an overall £200 million funding announcement that will support more than 1,000 new research and business leaders in AI across the UK.

References

^1. https://www.bbc.co.uk/news/science-environment-46046264

^2. https://www.cam.ac.uk/research/news/machine-learning-used-to-predictearthquakes-in-a-lab-setting

^3. https://www.cam.ac.uk/research/news/drones-used-toanalyse-ash-clouds-from-guatemalan-volcano

^4. https://www.cam.ac.uk/research/spotlight-on/artificial-intelligence

This article has been reproduced under a Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/)