Supplementary Code and Data for "Algorithm-Driven Root Optimization for Maize Yield in the Midwest"

Effective soil root exploration is central to improving water uptake and yield in maize. Using the Cycles Agroecosystem Model, we evaluated root distributions that maximize average yield by changing the water uptake amount and timing across the U.S. Midwest. Results revealed spatial patterns driven by climatic gradients. In the wetter eastern and central Midwest, a balanced root distribution with depth that can access incoming precipitation and subsoil water was favored, showing up to 33% higher transpiration and 26% higher yield than non-optimized phenotypes. The depth at which 50% of the roots are located (D50) was 0.8 m. In the drier regions along the western edge of dryland maize production and in transitional zones towards the east, bimodal root distributions which maintained access to both incoming precipitation and subsoil water prevailed, with the balance between a very top-heavy root distribution (D50 < 0.25 m) with a slight subsoil peak and a bottom-heavy root distribution (D50 > 1.25 m) with a medium surface peak dependent on the relative importance of surface vs subsoil water supplies. These locations had more consistent yields across years with bimodal root phenotypes. Phenotypes with reduced transpiration in these locations did not always penalize yield as water uptake shifted towards the reproductive phase, a key component of this unique adaptive strategy. This research highlights the vast potential of tailoring root architecture and proliferation with depth to regional hydric regimes to enhance resilience to climate variability and increase productivity in the Midwest and globally. This data describes the results obtained from optimization and subsequent analyses, along with source R code required to perform all analyses.