Integrating UAV and Ground-Based Hyperspectral Remote Sensing to Evaluate Split Nitrogen Application Strategies in Durum Wheat

Nitrogen (N) is one of the most important nutrients influencing wheat growth, plant nutrition, and grain production. Appropriate timing of nitrogen application is essential to synchronize nutrient availability with crop demand. This study evaluated seven nitrogen management treatments, including a control (N0) and six split nitrogen application schedules (N1–N6), in durum wheat under Mediterranean conditions using an integrated approach combining ground-based hyperspectral sensing and unmanned aerial vehicle (UAV)-based multispectral imagery. Plant nutrient concentrations (N, P, K, Ca, and Mg), spectral reflectance, vegetation indices, plant height, and grain yield were evaluated at different phenological stages. Split nitrogen application significantly affected plant nutrient concentrations, spectral reflectance, vegetation indices, plant height, and grain yield. Plant nutrient concentrations generally declined with crop development, whereas spectral reflectance increased across the visible and near-infrared regions of the spectrum. Vegetation indices derived from both hyperspectral and UAV multispectral data successfully differentiated phenological stages and nitrogen treatments. UAV-derived plant height showed strong agreement with field measurements, confirming the reliability of photogrammetric measurements for monitoring crop development. Among the nitrogen treatments, the N3 split application schedule produced the most favorable overall crop response, with higher plant nitrogen concentration, stronger spectral responses, and the highest grain yield. In addition, UAV-derived NDVI measured at the booting stage showed the strongest relationship with grain yield (r = 0.717, p < 0.01). These findings demonstrate that integrating ground-based hyperspectral sensing with UAV multispectral imagery provides complementary information for evaluating crop development and plant nutritional responses under different split nitrogen application schedules.