As part of its matchmaking approach that devises data-driven early action food security solutions for dry areas, the Jameel Observatory supports researchers looking for answers that pastoral communities can use to overcome climate changes. Observatory Students combine their academic studies with field work to produce development-oriented resources with practical applications.
Michael Renfrew is a PhD student at the University of Edinburgh’s School of Mathematics, and the Satellite Data in Environmental Science (SENSE) Centre for Doctoral Training – a UK partnership comprising the Universities of Leeds and Edinburgh, the National Oceanography Centre and the British Antarctic Survey. He joins after completing bachelor’s degrees in Medical Sciences at the University of Oxford and Mathematics and Statistics at the University of Edinburgh.
He brings an understanding of the application of mathematical modelling and statistical machine learning techniques to complex real-world problems.
Research focus and plans
Michael’s research uses machine learning and spatial regression models for agricultural livestock populations, trained on publicly available Earth Observation data such as Sentinel-2 hyperspectral imagery, very-high resolution (VHR) imagery from the wider project and unmanned aerial vehicle (UAV) data, ground-truthed in primary agricultural survey data.
Supported by the Natural Environment Research Council and the Food and Agriculture Organization of the United Nations, these will assist in downscaling the spatio-temporal resolution of agricultural livestock population estimates in East Africa, with an eventual view to augment expert opinion in the demand component of the Global Livestock Environmental Assessment Model (GLEAM).
Jameel Observatory significance
Supervised by Bruce Worton, Gary Watmough and Alan Duncan at the University of Edinburgh, Tim Robinson of FAO and Andy Challinor at the University of Leeds, Michael’s research will build on existing work on feed availability for livestock to enable more accurate and localised prediction of agricultural greenhouse gas (GHG) emissions and augment local resilience, planning and early warning systems, for instance in response to changing meteorological and climatological conditions.
Contact us for more information or to partner with us: