Implications of water management on methane emissions and grain yield in paddy rice: A case study under subtropical conditions in Brazil using the CSM-CERES-Rice model

Rice (Oryza sativa L.) is a staple food and plays a crucial role in the food security of many countries. However, rice cultivation is associated with significant methane (CH₄) emissions, contributing to overall greenhouse gas emissions and, thus, climate change. In this context, process-based crop models are useful tools for understanding and predicting the complex interactions between crop production, environmental factors, and sustainability. The objective of this study was to evaluate the performance of the Cropping System Model (CSM)-CERES-Rice model and DSSAT-GHG module to predict daily methane emissions and rice grain yield for different irrigation practices in a subtropical environment. The study employed a comprehensive approach, including measurements of daily CH₄ emissions, phenological stages, final aboveground biomass, and grain yield for rice cultivars BRS Pampa, BRS Pampeira, A705, and XP113 conducted over four consecutive crop seasons (2019–2023) and two irrigation systems: continuous flooding (CF) or alternate wetting and drying (AWD) in Capão do Leão, RS, Brazil. We followed a four-step methodology involving initial calibration of cultivar parameters, sensitivity analysis (soil-related parameters associated with CH₄ emissions), final cultivar parameters calibration, and long-term simulation analysis. Based on the sensitivity analysis and comparison to observed emissions, modifications were made to soil-related parameters such as soil buffer regeneration after drainage events (BRAD) and the fraction of soil water-filled porosity above which methane production occurs (WFPS_thresh) to enhance the accuracy of methane production. Optimal parameter combinations (WFPS_thresh = 70 %, BRAD = 0.070 d⁻¹) were selected based on a comparative analysis, enabling CH₄ simulations under non-flooded conditions. The predictive capability of the CERES-Rice model exhibited an average bias for grain yield of 485 kg ha⁻¹ under CF and 592 kg ha⁻¹ under AWD conditions. The results showed that the GHG module of DSSAT, after BRAD and WFPS_thresh parameter adjustments, was able to simulate daily CH₄ emissions in paddy rice with a very good agreement (average index of agreement (D-Statistic) of 0.87 for CF and 0.70 for AWD). Following the model evaluation, long-term simulations for different irrigation practices revealed the impact on grain yield, cumulative methane emissions, and seasonal applied irrigation. The highest crop water-methane productivity (CWMP = 52 %) was observed under sprinkler irrigation at 50 % soil water depletion, identifying it as the most sustainable option in this subtropical environment. Thus, the CSM-CERES-Rice model combined with the DSSAT-GHG module proved to be a potential tool for agricultural and environmental management of rice fields under subtropical conditions.