Introduction

The effects of climate change and environmental instability constitute some of the most pressing issues facing our society – affecting a large-scale interconnected system spanning nature, society and economic infrastructure. This has wide-ranging consequences for the local population, natural ecosystems and global climate. High-quality, accessible, and reliable information from different sectors can help build a comprehensive view of these complex issues. Modern data science techniques have already advanced our ability to integrate data and methodologies to find sustainable solutions from across all sectors. AI and Machine Learning (ML) can further deepen our understanding of climate change and inform evidence-based decision-making across a suite of policy areas.

AI for Science and Government (ASG) is a 38.8 million programme established at The Alan Turing Institute through UKRI’s Strategic Priorities Fund in 2018. ASG research brought together several organisations and experts to build national leadership in data science and AI, developing innovative techniques for fundamental and applied sciences as well as humanities research. Delivered in partnership with the Engineering and Physical Sciences Research Council (EPSRC) and in collaboration with several other research councils, ASG’s five-year vision was to demonstrate, via a diverse range of activities organised and presented as an integrated whole, how AI can be applied to effectively address urgent societal challenges and transform environmental research, engineering and government.

The Environment and Sustainability theme, one of three overarching areas of research in ASG, researchers undertook the role to understand and address one of the defining crises of our time — the climate emergency. This cross-theme collaborative initiative has enabled the development of AI technologies combining scientific, industrial, public and governmental efforts in environmental research. Under this theme, ASG has funded several initiatives led through multi-stakeholder partnerships, with the most significant strategic investments made towards the Data Science for Science and Humanities (DS4SH), Data-Centric Engineering (DCE) and Tools, Practices and Systems (TPS) programmes. DS4SH and DCE have evolved into large research programmes, instrumental in extending AI research at the institute and fostering an extensive portfolio of work in environment and sustainability. Experts are developing tools, methods and models to draw meaningful insights from data from across disciplines, improving risk management and guiding our response to climate change. TPS was established as a cross-disciplinary research programme offering leadership in open research and reproducibility by enabling inclusive collaboration and deploying interdisciplinary computational frameworks for managing the integration of data, methodology and expertise from different fields and sources. The long-term goals of these programmes will build on the legacy of ASG, shaping national and international strategies for building a sustainable future. ASG is supported entirely by public funds, through the UK Research and Innovation Strategic Priorities Fund, under EPSRC Grant EP/T001569/1 and EPSRC Grant EP/W006022/1.

In this report, we discuss projects involving data, software and methodologies under the Environment and Sustainability theme of ASG. Underpinning these ambitious efforts to tackle our greatest environmental and climate challenges are the processes for reproducible, ethical and collaborative research in AI and data science. Ensuring meaningful participation from diverse stakeholders has remained fundamental in different stages of the data life cycle, from planning and conducting research to developing methods, tools and technology. The Turing’s community building and convening capacity have contributed greatly towards strengthening interdisciplinary AI research and communities, enabling cross-domain relevance and applications by diverse user groups.