Environment and sustainability

Using data science and AI to tackle environmental and climate challenges


The effects of climate and environmental change constitute some of the most important challenges facing our society and quality, accessible, and reliable information across multiple scales, from global to molecular, is needed to understand the scale of these challenges and to develop strategies for mitigation and adaption.

Developing methods to provide the meaningful insight to inform decision-making, improve risk management and enhance our resilience to climate change will require working across disciplines, bringing together methodology and expertise from different fields to develop tools and computational frameworks that can integrate data from multiple sources, available at different spatial and temporal resolutions and with different biases and uncertainties.

The resulting frameworks will be applicable across a wide variety of applications, including integrating data from remote sensing satellites and traditional methods for ground observation with that from new sensor technologies; and integrating; integrating crop models, measurement data and climate projections to develop strategies for future agriculture; and understanding the threedimensional structure and dynamics of molecules.

This project is supported entirely by public funds, through Wave 1 of the UK Research and Innovation Strategic Priorities Fund, under EPSRC Grant EP/T001569/1.


Project aims

The programme has a number of aims including:

  1. Developing a framework to integrate earth observations from remote sensing satellites with that from ground sensors that acknowledges differences in data quality and resolution.
  2. Developing an integrated national crop modelling framework, using both models and data to allow the testing and development of new policies or management practices prior to implementation.
  3. Developing approaches to advance high-resolution determination of molecular structure and dynamics, enabling the understanding of living systems from molecular to cellular scales and the development of new materials.
  4. Development of generalisable tools and methods that can be used across a wide variety of applications.

Explaining the science

The programme combines inter-disciplinary expertise, covering a variety of areas including data science, AI, agriculture, environmental science, policy, meteorology, and ecology, from across the Turing community and external partners. Partners in this work include British Antarctic Survey, Earlham Institute, Norwich Bioscience Park (John Innes Centre), Rothamsted Research, The Met Office & Joint Centre for Excellence in Environmental Intelligence, University of Cambridge, MRC Laboratory of Molecular Biology and The Science and Technology Facilities Council.

The environment and sustainability programme currently consists of three projects:

  1. Environmental monitoring: blending satellite and surface data – This project aims to lay the foundations for ambitious research programmes to tackle our greatest environmental changes. We will develop and deploy reproducible and interpretable methods to increase scientific understanding, build tools to help environmental measurement planning, and provide the underpinning tools for intelligent real-time monitoring. The project aims to strengthen collaborations across the environmental and machine learning/AI communities by enhancing open-source and user-friendly computing platforms, including Pangeo.
  2. The impact of climate change on agriculture – Agriculture is highly vulnerable to climate change and it is likely that in fifty years’ time the crops we currently grow in the UK will no longer be viable to meet consumer demand. Agriculture also has important socio-economic implications for food sustainability and greenhouse gas emissions. We urgently need to begin investigating alternative strategies for agriculture that will require bringing together the best available models for the key aspects of the most important UK arable crops. This project will pioneer the development of methodology that will integrate data from plant science, hydrology, soil science, insect population dynamics, economics, consumer behaviour and climate models to form an integrated national crop modelling framework to support new policy or management practices, by offering simulations of their impact under different climate change scenarios.
  3. Molecular structure from images under physical constraints – The determination of high-resolution molecular structures is critical to our fundamental understanding of living systems, and for the development of novel drugs and materials. There are several outstanding challenges. First, we need to develop approaches that will allow us to intelligently identify candidate structures from image data, and tools to generate the large amounts of annotated training data required for machine learning models. Second, we need to reconstruct molecular structure from the observations by leveraging advances in joint reconstruction and deformation models. Finally, we need to incorporate physical constraints as a prior during the determination of molecular structure. Adding methods to cope with dynamic and continuously changing structures would be a considerable advance. The project aims to develop new machine learning/AI methods to address these challenges, and to advance the state-of-the-art in molecular structure determination.

Work in the environment and sustainability area addresses one of the defining crises of our time: the climate emergency. Developing robust and complex models integrating multiple datasets, real-time monitoring and data sources will enable enhanced decision making across a suit of policy areas. Sophisticated molecular modelling techniques will unlock more advanced and sustainable materials. Researchers will also be developing a toolbox of generalisable tools that will open the advances made to other areas of research, compounding the long-term potential impact of the work.