Circulation types and AI for crop yield anomaly detection and prediction

Climate change has the potential to create volatility in our food supply chain, particularly in key crops such as winter wheat. This research examines how large-scale weather patterns can be utilised to enhance medium to long-range predictions of wheat yields in the UK. By combining the occurrence and persistence of weather patterns with machine learning models, the study demonstrates that it's possible to forecast yields months in advance of harvest. The findings also reveal how regions of the UK may be affected differently by future climate scenarios, with some areas seeing yield increases and others declines. This work presents a novel approach to enhancing crop forecasts' accuracy and timeliness, enabling farmers and policymakers to prepare for a changing climate.