International Journal of Plant Production Agronomy and Crop Science
Simulating the Response of Rice Production and Nitrogen Uptake to Different Nitrogen Application Levels and Splitting Strategies Using the Process-Based CERES-Rice Model
Abstract
Nitrogen (N) management is a key factor in achieving optimal rice production globally. Beyond improving grain yield, nitrogen application rate and splitting strategy strongly influence biomass accumulation, canopy development, and nitrogen uptake dynamics throughout the rice growth cycle. However, reliable simulation of these responses across contrasting N management strategies remains a challenge for process-based crop models. This study aimed to calibrate and validate the CERES-Rice model for simulating grain yield and N uptake of rice (Shiroudi variety) under various N amount and splitting treatments in the Amol region, Iran. A four-year (2016–2019) factorial field experiment was conducted. Data from the first two years (2016–2017) with five N levels (0–160 kg ha⁻¹) and four splitting patterns were used for calibration, and data from 2018 to 2019 (0–180 kg ha⁻¹ and four patterns) were used for validation. Model performance was evaluated using the coefficient of determination (R²) and normalized root mean square error (RMSEn). The CERES-Rice model demonstrated excellent accuracy in simulating grain yield, with R² values ranging from 0.82 to 0.90 and RMSEn values between 3.4% and 7.0% across both phases. The results further showed that moderate to high nitrogen levels (80–180 kg ha⁻¹), particularly when applied through balanced split applications (S2 and S3), enhanced biomass production, improved nitrogen uptake, and resulted in the most accurate yield simulations. These nitrogen strategies synchronized N availability with crop demand during tillering and panicle initiation stages, thereby improving yield formation and crop growth dynamics. The simulation accuracy for biomass and total N uptake was “good” to “moderate” (RMSEn = 9.1–13.2%), but lower for Leaf Area Index (LAI) (RMSEn up to 21.6%). The model consistently captured the positive relationship between increasing nitrogen supply and total plant nitrogen accumulation, while also reproducing the beneficial effects of split nitrogen applications on grain nitrogen concentration and biomass accumulation. However, under N-deficient conditions, the model tended to overestimate grain yield and nitrogen uptake, indicating limitations in simulating severe nitrogen stress and associated source–sink restrictions during early crop growth. The results confirm that the calibrated CERES-Rice model is a reliable decision support tool within the DSSAT framework for optimizing N management and splitting strategies for sustainable rice production in the Amol region. Overall, the findings emphasize that synchronizing nitrogen application timing with crop developmental stages is essential for maximizing nitrogen use efficiency, biomass accumulation, grain yield, and nitrogen uptake in rice-based production systems.