Data scientists from across the UK gathered at Lancaster University last week for the first of what's hoped to be the annual UK Conference on Environmental Data Science (https://wp.lancs.ac.uk/ceds/). The meeting had four key themes: data science infrastructure, net-zero, extremes and resilience.
An Early Career Researcher (ECR) day, which included a tour of the historic Lancaster castle, kicked the week off. We experienced exciting perspectives from academics and industry partners, including a panel discussion on the merits and disadvantages of a career in the respective fields. Adding to the dilemma facing ECRs, we had industry professionals speaking about the draw of academia while senior academics highlighted the appeal of industry. ECRs had the chance to mingle and share ideas in a way that's been difficult for the last two and a half years. For many in attendance, it was the first in-person conference of their PhD!
The remainder of the conference consisted of Plenary sessions detailing future forests (Data and data science for predicting the future of forests, Emily Lines), the importance of metadata (Thrashing feet, gliding swan – metadata challenges for environmental data science, Lucy Bastin) and our approach to achieving net-zero (Getting to Net Zero, Emily Shuckburgh). We had an inspiring presentation from Suman Ravuri and Kevin Donkers on a DeepMind-Met Office collaboration to model precipitation nowcasting (Skilful precipitation nowcasting using deep generative models of Met Office radar data).
The first hands-on sessions described machine learning approaches and introduced efficient means for data loading, handling, exploration and visualisation. This follow-along session included developing a demo ML pipeline (Introduction to Machine Learning, Stephen Hadad). The second hands-on session detailed the intricacies of GIS programming (Stop Clicking and Start Coding – GIS Programming in R, Barry Rowlingson).
During short and snappy, two-minute PICO presentations, attendees had the chance to highlight their work to interested parties. The main poster sessions facilitated in-depth conversation. The projects on display covered aspects of the natural environment: ice caps and oceans, air quality and ozone, forests and croplands. Many of the techniques on offer exploited machine and deep learning algorithms in some highly informative presentations.
In attendance, we had representatives from Lancaster University, Exeter University, British Antarctic Survey, UK Centre for Ecology and Hydrology, Deep Science Ventures, Oxford University, Glasgow University, The Alan Turing Institute, Conflict and Environment Observatory, University of Cambridge, Plymouth Marine Laboratory, University College London, University of Bristol, British Geological Survey, Aston University, DeepMind and the Met Office.
The opinions expressed by our bloggers and those providing comments are personal, and may not necessarily reflect the opinions of Lancaster University. Responsibility for the accuracy of any of the information contained within blog posts belongs to the blogger.
Back to blog listing