Agroecosystems produce food but also deliver both benefits and disbenefits for the environment, including water and air quality. Climate change forecasts for the UK suggest changes to the intensity and frequency of flooding, drought, heat and strong winds, as already evidenced by the record-breaking years observed in the UK over the past decade. At present, we do not have reliable data on the impacts of climate stresses – including extreme events – on key crops (e.g., cereals and grass). This means recommendations for cereal and livestock farm management strategies to help maximise resilience to future climate change cannot be made with confidence. As a result, 47% of farmers feel ‘not at all positive’ about their future in farming.
This PhD will develop a novel ‘Big Data’ analytical ‘toolbox’ consisting of statistical and machine learning tools to detect extremes and anomalies in data collected at Rothamsted’s North Wyke Farm Platform. Now in its 12th year of operation the platform’s database consists of over 70 million measurements on all major inputs and outputs for four different farming systems. Detection and accounting for extremes is of great significance in advancing our understanding of how climate change impacts different agroecosystems. The analytical ‘toolbox’ for capturing ‘Out of the Ordinary’ will also strongly contribute to planned Digital Twins of the farm platform.
For the successful student, considerable training in a range of data and digital science skillsets will be provided. This includes those in high performance computation, advanced statistics and machine learning, together with specialisms in extreme-value theory, anomaly detection and the analysis of spatio-temporal processes. The farm platform experiment is world-leading where methodological advances to the analysis of its rich datasets should lead to several high-impact publications and makes a formidable opportunity to start a research career.
Applicants should hold a minimum of a UK Honours Degree at 2:1 level or equivalent in subjects such as Environmental Science, Applied Statistics, Geography, Data Science, Biology, Natural Sciences, Computer Sciences or related disciplines. Experience of working with big data, spatial databases (e.g., GIS) and writing computer code (e.g., in R, python) would be advantageous, although training will be provided.
For further information or informal discussion about the position, please contact Prof Paul Harris (email@example.com).