LEEDS MINDS, GLOBAL IMPACT
Thanks to support, Leeds researchers are finding solutions to challenges faced by communities around the globe.
Forecasting a safer future
In Western Africa, increasingly unpredictable weather patterns have made forecasting challenging, often resulting in poor crop yields and economic instability.
Farmers’ livelihoods depend on determining when wet seasons begin, and when dry or wet spells are on the way – and therefore when to plant, irrigate, weed, prune, apply fertilizer, pesticides, and harvest crops. Getting it wrong can have damaging consequences for food security – and the impacts of climate change are making this increasingly likely.
The traditional forecasting methods designed in Europe and the US are less effective in Sub-Saharan Africa, where the weather behaves differently. But working alongside the Alan Turing Institute, Cambridge University and a consortium of partners, Leeds researchers are working to change the story.
Project Cumulus, funded by the Gates Foundation and the Foreign, Commonwealth and Development Office, will develop AI-based forecasting methods tailored to African conditions which will deliver more accurate and locally relevant insights. It will combine satellite imagery, ground observations and existing forecast data to create a clearer picture of the atmosphere, which can help farmers improve crop yields and reduce economic losses.
“AI allows us to make more accurate predictions of the probability of rainfall,” says Douglas Parker, Professor of Meteorology in the School of Earth and Environment. “The technology will learn from observations in each country and bring in information from weather and ocean patterns around the globe.
By working with top universities and weather services in West Africa, this international partnership will drive innovation in AI weather prediction.”
Sink or swim
The Congo Basin is the world’s second largest tropical rainforest, and home to the world’s largest peatland complex. But with processes such as deforestation, degradation and regrowth of forestland under way, it is uncertain whether this crucial part of the world’s surface is slowing climate change – or actually accelerating it.
Using advanced modelling, and aircraft and satellite data, Leeds researchers will take the first ever airborne measurements of carbon dioxide (CO2) in the Congo Basin to understand its role in the global carbon cycle.
“The world’s land surface plays an essential role in the global carbon budget, but there are major data gaps in this area of research,” says Simon Lewis, Professor of Global Change Science. “The Congo Basin is a particularly under-studied region.”
The Congo-FLEX project brings together partners across the US, UK, Europe and Africa to assess whether Central Africa’s forests are acting as a carbon sink by absorbing CO2 from the atmosphere, or if they are releasing it – and why. It will help inform policy decisions and provide early warnings for action.
This partnership is one of four research teams drawn from different organisations that have been awarded up to £33 million over five years to advance scientific understanding of the carbon cycle and its influence on a changing climate. The teams are the first to be funded through Schmidt Sciences’ Virtual Institute for the Carbon Cycle, a philanthropic organisation working towards a healthy, resilient, secure world for all.
Using AI to reduce food waste
Leeds researchers are developing an AI tool to help transform food waste into valuable protein.
Each day, hundreds of tonnes of agri-food waste is generated around the globe – often from damaged crops that have not been picked. It contains high-quality protein, which could be upcycled for use in standard food products or to produce meat and dairy alternatives. But to obtain it requires a complex fermentation process, with optimum fermentation conditions hard to achieve due to the variable nature of the waste. The result is a product that ends up being more expensive to produce than the non-upcycled equivalent, making adoption unlikely in the food sector.
“To truly impact global food security, upcycled protein can’t just be a niche alternative,” says Nicholas Watson, Professor of Artificial Intelligence in Food. “It has to compete on price with what is already on the supermarket shelf. That’s the gap we’re trying to bridge.”
Funded by the Bezos Earth Fund, Leeds academics will lead a global team of experts to create a tool that can be used to calculate the optimal set of fermentation conditions to produce microbial protein for the lowest possible cost. They will help industry to develop proteins tailored to their needs at a cost comparable with non-upcycled options.
Built with Shorthand