Leaf to Root: Harnessing leaf spectral signatures for non-destructive monitoring of soybean nodule traits

Soybean ( Glycine max ) root nodules, formed through symbiosis with nitrogen-fixing rhizobia, are essential for biological nitrogen fixation. While quantifying key nodulation traits, nodule number and weight, is critical for assessing symbiotic efficiency and yield potential, current methods are destructive and labor-intensive, unsuitable for longitudinal monitoring and high-throughput phenotyping. Here, we established hyperspectral leaf reflectance as a non-destructive, high-resolution tool capable of monitoring root nodule development. Using Partial Least Squares Regression models, we connected spectral data with nodule metrics from 528 unique soybean plants across 18 genotypes, inoculated with different rhizobium strains, and under different abiotic stresses. These models achieved high accuracy for predicting nodule number (R 2 = 0.75, nRMSE = 6.02%) and moderate accuracy for nodule weight (R 2 = 0.53, nRMSE = 12.38%). Crucially, spectral analyses revealed distinct hyperspectral signatures sensitive to nodule traits. While different rhizobium strains induced comparable changes in both nodule traits, and therefore produced highly overlapped spectral domains, diagnostically distinct spectral patterns were generated under drought versus salt stress, with the former suppressing nodulation more significantly than the latter. Furthermore, we demonstrated the effectiveness of our models for real-time in-situ monitoring of nodule development for individual plants. Spectral-nodule trait covariation analyses further revealed leaf signatures correlated with nodule traits primarily through systemic physiological coupling governed by carbon-nitrogen exchange dynamics and plant water status. This study showcased hyperspectral sensing as a transformative methodology, enabling the unprecedented non-destructive quantification of nodulation dynamics, revealing novel physiological insights into plant-microbe-environment interactions, facilitating breeding and management strategies for sustainable soybean production.