Submitted by Rachel Gardner on Mon, 29/01/2024 - 15:30
We shone the spotlight on our research in Climate and Sustainability in January 2024.
Our Research Showcase on Wednesday 24 January 2024 featured a series of short talks by early-career researchers from this Department and from the AI for the study of Environmental Risks (AI4ER) doctoral training centre, which we co-host.
Amelia Holcomb is one of the researchers who recently developed an algorithm that uses computer vision techniques to accurately measure trees almost five times faster than traditional, manual methods. In her talk on 'Space Lasers for Good: Remote measurement of tropical forest degradation', she described her work on combining spaceborne light detection and ranging (LiDAR) with other sensors to provide large-scale, high-resolution measurement of the carbon emissions associated with tropical forest degradation.
PhD student Patrick Ferris talked about 'Uncertainty at scale: how computer science hinders climate research'.
"Computer science is a powerful tool for enabling data-driven advances in global ecology and conservation," Patrick says. "But the amplification cuts two ways, as mechanisation can also compound problems inherent with just how uncertain anything to do with natural ecosystems are! In this short talk, I look at the different ways computer science amplified uncertainties in our work to analyse avoided deforestation projects in tropical moist forests."
Petr Dolezal's talk had the intriguing title 'How old weather forecasts can help us design power systems of the future'. Petr is co-supervised by Emily Shuckburgh, our Professor of Environmental Data Science and Srinivasan Keshav, Robert Sansom Professor of Computer Science here.
Petr is repurposing expired ensemble weather forecasts, harnessing the chaotic nature of weather to generate over 10,000 years of independent data. "This expansive dataset enables a comprehensive exploration of rare and extreme weather scenarios and aids in designing robust renewable systems resilient to weather variability," he says. "It also presents challenges that can be addressed by careful systems design and distributed computing."
Anais Berkes has studied AI’s role in the transition to clean energy and will talk on 'The Optimal Design of PV-EV Integrated Domestic Microgrids: A Post-Pandemic Perspective'. She discussed how the post-pandemic increase in remote working has led to electric vehicles being more frequently plugged in at home, enabling their use as bi-directional energy storage units within photovoltaic-powered microgrids. And she proposes a novel algorithm to jointly size and operate these microgrids.
Simon Mathis is interested in using data science to mitigate climate change and its effects and talks about 'AI designed biotechnology to tackle environmental challenges'. He highlights several areas where biomolecules can make a tangible difference to environmental challenges, ranging from plastic degradation to replacing petrol-based chemistry through biochemistry. And he discusses how AI can help design and improve biomolecules that enable such feats.
