EVALUATING THE PERFORMANCE OF THE CERES-MAIZE MODEL FOR DECISION-SUPPORT IN NIGERIAN MAIZE PRODUCTION SYSTEMS

Abstract

The study focused on evaluating the performance of the CERES-Maize model within the DSSAT framework for simulating the growth and yield of three maize varieties (2016 TZdEE-Y, 2014 TZE-Y, and SAMMAZ-40) in the savanna region of Nigeria. Field experiments were conducted during the 2023 growing season across three locations in Kano state, representing both the Sudan and Northern Guinea Savannas. The experimental data, which included six different sowing dates, were used to validate the model. The results demonstrated that, the calibrated CERES-Maize model achieved a high degree of accuracy in replicating observed data for the tested varieties. Key performance indicators, including a d-index ranging from 0.75 to 0.98 and an nRMSE of 1.6–8.96% for traits such as days to anthesis, physiological maturity, grain yield, and total dry matter, confirmed the model's robustness. This evaluation successfully validates the CERES-Maize model as a reliable tool for simulating the phenology and productivity of these specific maize varieties under the diverse sowing conditions of Nigerian savannas.