Modeling organically fertilized flooded rice systems and its long-term effects on grain yield and methane emissions

Abstract

The increasing trend of adopting organic fertilization in rice production can impact grain yields and soil methane (CH4) emissions. To simulate these impacts in the absence of long-term field data, a process-based biogeochemical model, Denitrification and Decomposition (DNDC version 9.5) was used. The model was calibrated against a single year greenhouse study and validated using a previously published one-year field trial from 1990, both comparing varying fertilization systems in rice production in southeast Texas, USA. In both the greenhouse and the field studies, lower grain yield and greater soil CH4 emissions were observed in organically fertilized systems. Calibrated model simulations of the greenhouse study correlated with the observed daily CH4 emissions (conventional r2 = 0.87; organic r2 = 0.91) and SOC (r2 = 0.83); but, the model overestimated yield of conventional systems (slope = 1.2) and underestimated yield of organic systems (slope = 0.68). For the field study, agreement between simulated and observed yields and CH4 emissions resulted in slopes close to 1. A simple organic system with urea and straw amendment from the field study was an input available in DNDC whereas the slow release, pelletized organic fertilizer used in the greenhouse study, Nature Safe, was not modeled well by DNDC. The validated model was used to simulate 22 years of rice production and predicted that the differences in yield and CH4 emissions between treatments would diminish with time. In the model simulations, the overall soil health was enhanced when managed with organic fertilization compared to conventional inorganic fertilizers. Model simulations could be improved further by including site-specific calibration of soil organic C, and soil carbon dioxide (CO2) emissions.