We present a computational framework that integrates functional-structural plant modeling (FSPM) with an evolutionary algorithm to optimize three-dimensional maize canopy architecture for enhanced light interception under high-density planting. The optimization revealed an emergent ideotype characterized by two distinct strategies: a vertically stratified leaf profile (steep, narrow upper leaves for penetration; broad, horizontal lower leaves for capture) and a radially tiled azimuthal arrangement that breaks the conventional distichous symmetry of maize to minimize self and mutual shading. Reverse ray-tracing simulations show that this architecture intercepts significantly more photosynthetically active radiation (PAR) than virtual canopies parameterized from high-performing field hybrids, with gains that generalize across multiple U.S. latitudes and planting densities. The optimized trait combinations align with characteristics of modern density-tolerant cultivars, supporting biological plausibility. Because recent gene editing advances enable more independent control of architectural traits, the designs identified here are increasingly feasible. By uncovering effective, non-intuitive trait configurations, our approach provides a scalable, predictive tool to guide breeding targets, improve light-use efficiency, and ultimately support sustainable yield gains.
@article{arxiv.2512.06064,
title = {Towards smart canopies: Algorithmic design of maize canopy architectures that maximize light use efficiency},
author = {Nasla Saleem and Talukder Zaki Jubery and Yan Zhou and Yawei Li and Adarsh Krishnamurthy and Patrick S. Schnable and Baskar Ganapathysubramanian},
journal= {arXiv preprint arXiv:2512.06064},
year = {2025}
}