Reflectance-based assessment of nitrogen status in ryegrass and mixed ryegrass-clover intercropping fodder crops

Abstract

Effective nitrogen (N) management is essential for optimizing crop yields and minimizing environmental impacts, particularly in semi-arid regions where climate risks and natural resource constraints complicate decision-making. These low-energy systems require precise N strategies tailored to their unique challenges. This study evaluated a sensor-driven data analysis workflow for assessing N status in ryegrass-based fodder crops under semi-arid conditions and identified the most effective bands and vegetation indices (VIs) for use. Field trials conducted at Herdade da Comenda in Portugal employed a split-plot design, testing three N topdressing rates (0, 120, and 200 kg ha⁻¹) across varying crop types and irrigation systems. Both physical and remote measurements of crop parameters and N nutrition indicators were taken to address the limitations of current approaches in these conditions. The study found that vegetation pixels dominate spectral imagery, making additional filtering, such as ExG masks, unnecessary at ryegrass tillering and stem-elongation in ryegrass-based fodders. This simplification reduces processing time, costs, and digital footprints. Key VIs—NDRE, RERVI, and CIRE—proved robust for monitoring variables such as crop type, growth stage, and N treatments, showing strong correlations with N status indicators (NNI and CNI). Additionally, the study contrasted the efficiency of the entirely remote NNI method with the enhanced accuracy of the hybrid CCCI-CNI approach, providing valuable insights for tailored N management in semi-arid systems.