From Simulation to Discovery: AI Enabled Probabilistic Emulation of Mechanistic Crop Systems
Researchers have developed a novel AI-powered probabilistic emulator for crop modeling, significantly reducing computation time by several orders of magnitude. This emulator, trained on millions of simulations and augmented with a synthetic weather generator, allows for scalable exploration of crop responses under diverse environmental conditions. The framework has been applied to identify maize trait combinations that maintain high yields across various scenarios and has revealed that radiation use efficiency and root dynamics are key drivers of yield resilience. AI
IMPACT Enables large-scale discovery of crop trait combinations for improved yield resilience under climate change.
