Assessing maize growth and yield response to different P fertilizer levels with an adapted DSSAT CSM‐CERES‐Maize model

Abstract Phosphorus (P) is essential for maize ( Zea mays L.) growth but is a limited and non‐renewable resource. Studies using DSSAT CSM‐CERES‐Maize remain scarce for low‐P soils. This study assessed maize yield response to P fertilization in P‐deficient soils using an adapted version of the DSSAT CSM‐CERES‐Maize model with minor code modifications to vegetative partitioning. The CERES‐Maize was calibrated and evaluated using data from two field experiments (2020–2023) conducted in Southern Germany on P‐deficient soils, with P fertilizer treatments of 0 and 85 kg P ha −1 in 2020 and 2021, and 0, 25, 50, and 75 kg P ha −1 in 2022 and 2023. Evaluation results showed a good agreement between simulated and measured data for aboveground biomass (normalized root mean square error [nRMSE] = 24.4%, d‐stat = 0.98) and grain yield (nRMSE = 11.4%, d‐stat = 0.99). While overall leaf area index simulations were satisfactory across treatments (d‐stat = 0.87), accuracy was lower in low or zero P fertilizer treatments, with a d‐stat of 0.72 for the zero P treatment compared to 0.97 for the 85 kg P ha −1 treatment in 2020 and 2021. CERES‐Maize can be used to support P management decisions, but it showed a limited ability to simulate leaf/P dynamics under very low P levels in this study. These findings confirm the capability of CERES‐Maize to simulate maize growth and yield responses to varying P fertilization. Future modeling research should further investigate performance under low P fertilizer levels in greater detail across diverse environments to enhance and validate the DSSAT P subroutine.