Assessing cropland suitability is vital for future agriculture as it optimizes land use, identifies high-potential areas, and supports sustainable food production while minimizing environmental impacts. However, projections of future crop yields based on suitability ratings remain debated due to the complexities of climate change, soil degradation, and evolving agricultural practices. In particular, yield projections that hinge on CO 2 fertilization are contentious. While rising CO 2 levels can enhance photosynthesis, studies indicate this effect is smaller than once believed and insufficient to offset climate change's negative impacts. Its benefits are further constrained by temperature, water, and nutrient limitations. Using a deep learning approach trained on historical soil, yield, and climate data to evaluate and predict multi-crop suitability, we generate new forecasts of Canada's cropland suitability for major annual crops in 2050 and 2100 (under RCP 4.5 and 8.5). Results show declining suitability for canola, peas, spring wheat, and soy in the Prairies, with gains for barley and oats. Similar shifts are projected in central British Columbia, north of Southern Ontario, and Southern Quebec. Net losses in canola and spring wheat are expected to outweigh gains in other crops, underscoring the need for adaptive management strategies such as crop diversification and the development of heat-resilient varieties.
Bhullar et al. (Fri,) studied this question.