Explainable AI shows climate impacts on wheat yields: insights from 30 years of field data.
Authors: Visse-Mansiaux, M., Ryo, M., Burton, A., Siraj, T., Schiller, J., Treier, S., Pellet, D., Stefan, L., Levy Haener, L., Herrera, J. M., Wanger, T. C.
Category: Plant Biology
Model Organism: Triticum aestivum
▶ AI Summary
The study leverages a 30‑year dataset of winter wheat trials across six Swiss sites, applying explainable AI and interpretable machine‑learning methods to identify key climatic drivers of yield. Findings highlight cumulative solar radiation, precipitation, and genotype composition as primary determinants, with a yield plateau observed above ~3000 MJ m⁻² of solar radiation, indicating complex genotype‑by‑environment interactions. The framework demonstrates how XAI can enhance biological insight and guide breeding and adaptation strategies under climate change.