On 22-23 March, the Alan Turing Institute is hosting Artificial Intelligence UK (AIUK). Broadcast live from London, this virtual event presents a showcase of the UK’s latest research into Artificial Intelligence (AI) and data science to explore how the pioneering work and collaboration taking place in these fields can be applied to solve real-world challenges.  

The Met Office is partnering the event, which will see panel sessions, demonstrations and talks from a range of thought leaders in the data science sector. Chief Met Office Scientist, Professor Stephen Belcher, took part in this morning’s panel session “AI and climate: a cause for hope?” in which speakers reflected on COP26 and the role of AI in the path to net zero and climate change adaptation over the next decade. 
 
AIUK comes at a time of pivotal growth for Artificial Intelligence, and an expansion in what is possible brought about by an explosion in data, increased availability across a broad spectrum of computing and growing expertise in data science approaches (including machine learning and AI).  An acceleration in the development and adoption of new data science presents an enormous opportunity, offering significant gains in efficiency and performance to most sectors, including but not limited to weather and climate science and services. 

Dr Kirstine Dale, Principal Fellow for Data Sciences at the Met Office and Co-Director for Joint Centre for Excellence in Environmental Intelligence said: “We’re in the midst of a revolution in Artificial Intelligence, where the world’s fastest-growing deep technology has the potential to rewrite the rules of entire industries, fundamentally changing the way we work and live. 

“At the Met Office we’re committed to harnessing the power of data science to push the frontiers of weather and climate science and services.” 

The Met Office works in partnership with a range of institutions to fulfil ambitions within data science. For example, the Joint Centre for Excellence in Environmental Intelligence was launched in 2020 with the University of Exeter, while the Met Office works collaboratively with the Alan Turning Institute, as well as leading universities, to further explore the applications of data science within weather, climate and AI.  

Kirstine continued: “Working at the cutting edge of this transformative technology requires collaboration, and that’s why we’re engaging with world-leading partners in the application of machine learning, data science and AI. 

“By working together with other organisations in this area, we’re able to fully explore and embrace different opportunities and challenges within data science.” 

One such challenge is space weather: by deploying machine learning and data driven techniques, the Met Office hopes to improve how earth system models model the thermosphere – the part of our atmosphere where many satellites fly. These techniques would enable improvements to the model in areas where either the science hasn’t been fully discovered or when the process is too computationally intensive to fully model and so must be approximated. 
 
Theo McCaie, Head of the Informatics Lab and Technical Lead for Data Science said: “The earth’s system is extremely complicated and events that happen lower in the atmosphere can affect the thermosphere and vice versa. However, modeling an atmosphere 600km thick all the way around the earth is scientifically challenging and computationally very expensive. 

“We are researching the use of machine learning techniques to emulate some of the atmospheric processes such as gravity waves. These emulations could be computationally much, much cheaper and possibly more accurate. Alongside this we are using data driven techniques to improve the existing model components to make them more accurate. Used together, these techniques could make modelling and forecasting in the thermosphere timelier and more accurate, leading to a safer, more reliable environment for the space-based components of many of the services we rely on day to day. 

“Space weather is just one example of the application of Data Science. We don’t know what the next advancements or opportunities will be in AI, but we’re making sure that we’ll be ready to take advantage of them, ensuring that we fulfil our purpose to help people make better decisions to stay safe and thrive.”