Mid-season variable-rate algorithm for organic and mineral rice fertilization

Abstract

Accurate calculation of nitrogen requirements is essential in rice fields utilizing both local manure and mineral fertilizers to mitigate nitrogen deficiencies and yield losses associated with reducing chemical fertilizer use. Traditional approaches often fail to effectively integrate organic and mineral fertilizers or adapt to the complexities of real-farm conditions. To tackle these challenges, this study introduced a novel application of the Nitrogen Fertilizer Optimization Algorithm (NFOA), leveraging remote optical sensors and Sentinel-2 satellite imagery to deliver precise, data-driven nitrogen recommendations for the effective integration of organic fertilization in rice cultivation. Fertilizer prescription maps generated by the NFOA delivered precise nitrogen recommendations tailored for diverse real-farm fields. The algorithm demonstrated strong predictive performance for yield responses to nitrogen application at critical phenological stages, such as panicle initiation and maximum tillering (R² = 0.71, p < 0.0001; R² = 0.73, p < 0.0001). Key findings demonstrate the model’s ability to optimize nitrogen inputs, achieving up to a 40% reduction in surplus nitrogen while maximizing yields. By promoting a balanced nitrogen input-output equilibrium, the NFOA offers significant environmental and economic benefits, even in the context of the complexities associated with organic fertilization. In conclusion, these findings suggest that the NFOA approach is suitable for calculating nitrogen fertilizer requirements in rice fields using organic fertilization strategies, effectively accommodating the high variability in nutrient content and availability of organic nitrogen to rice crops. However, further refinement is necessary to enhance its predictive accuracy by incorporating advanced spectral indices and accounting for detailed environmental and management factors.