Selection of non-destructively measured wheat traits sensitive to nitrogen application and closely related to wheat yields

Rapid and non-destructive detection of crop growth traits can guide nitrogen (N) diagnosis and management. However, there is a lack of comprehensive studies on multiple non-destructively measured crop traits under different N inputs. The study aimed to investigate the responses of 36 non-destructively measured wheat traits to N application rates, and to assess which traits are sensitive to N application and closely related to wheat yields. The 36 traits included plant shape traits, physiological traits, physical traits, and leaf color traits. Field experiment included 24 treatments, i.e., the combination of eight N application rates (0, 50, 100, 150, 200, 250, 300, and 350 kg N ha⁻¹) and three farmlands (farmland converted from wasteland for 2, 5, and 8 years, denoted as Farmland_2yr, Farmland_5yr, and Farmland_8yr). Results showed that the N application rate for the greatest grain yield was 350.0 (Farmland_2yr), 286.7 (Farmland_5yr), and 217.6 kg N ha⁻¹ (Farmland_8yr). Nitrogen application tended to increase most plant shape, physiological, and physical traits; while it tended to reduce most leaf color traits. Among the 36 traits studied, 19 traits were sensitive to N application and closely related to wheat yields. The 19 traits were plant height, stem diameter, mean leaf length, mean leaf width, mean leaf area, mean leaf volume, single stem leaf area, leaf area index, leaf SPAD, leaf quantum yield, leaf N content, leaf surface moisture, leaf surface temperature, canopy radiation transmittance, leaf NBI, RDBI, GDBI, BDBrI, and RDLI. These traits can be considered for the establishment of rapid N diagnosis systems in wheat production. The findings are expected to deepen our understanding of the responses of multiple non-destructively measured plant traits to N levels, and provide essential information for the establishment of rapid N diagnosis systems and grain yield prediction models.