Forecasting future crop suitability with microclimate data

Against a background of unprecedented climate change, humanity faces the challenge how to increase global food production without compromising natural environment. Crop suitability models can indicate best locations grow different crops and, in doing so, support efficient use land leave space for, or share with, nature. However, challenges downscaling data needed drive these make predictions for future has meant that they are often run using national regional projections. At finer spatial scales, variation conditions have substantial influence on yield and so continued coarse resolution risks maladaptive agricultural decisions. Opportunities novel crops, which knowledge local microclimate may be critical, missed. We demonstrate information acquired region used mechanistic crop model under present day possible scenarios. modelling techniques generate 100 m datasets south-west UK (2012–2017) predicted (2042–2047) time periods. WOrld FOod STudies (WOFOST) 56 varieties, returns maximum yields each planting month. Over short distances, we find highest attainable vary substantially discuss differences mean field-level assessments could land-use decisions, enabling whilst protecting biodiversity. provide code running WOFOST WofostR R package, thus integration with meaning our methodology applied anywhere world. As such, available anyone tools predict at high both