Breeding new high-yield and high-quality forage soybean cultivars is an effective approach to addressing the shortage of feed protein and sustainable agricultural development. Plant height is a key indicator of forage soybean genotypes and is closely related to forage yield. However, the determinants affecting soybean plant height remain highly uncertain. In order to analyze the factors contributing to plant height differences among soybean cultivars, two tall-stemmed soybean cultivars (“Neinong S001” and “Neinong S002”) and two short-stemmed soybean cultivars (“Neinong 0004” and “Neinong 0005”) were used in this study as test materials for examining aboveground phenotypic characteristics, root traits, and photosynthetic characteristics. The test materials were planted in 2018 at Chakintai Ranch (122°15′ E, 43°38′ N) using the potting method, and the indicators were measured in June. The results showed that the leaf area, root volume, and root surface area of high-stemmed soybean cultivars were significantly (p < 0.05) lower than those of short-stemmed soybean cultivars. Additionally, the dry weight of a single plant and transpiration rate were significantly (p < 0.05) higher in high-stemmed soybean cultivars compared to short-stemmed soybean cultivars. It was found that soybean plant height was significantly (p < 0.05) correlated with leaf area, leaf shape index, intercellular CO₂ concentration, transpiration rate, SPAD, root weight, root length, root surface area, and root volume. Further path analyses revealed that intercellular CO₂ concentration and root surface area had a direct impact on plant height, with direct effect coefficients of 0.22 and −0.91, respectively. These results provide new insights into the sustainability development and genetic enhancement of plant height characteristics in forage soybean.

The negative relationship between root yield and sugar content remains an important challenge for sugar beet (Beta vulgaris L.) improvement. Optimizing canopy architecture may provide an effective strategy for identifying germplasm with favorable combinations of productivity and quality. In this study, 216 sugar beet accessions were evaluated under a single planting configuration at the Liangcheng Experimental Station, Inner Mongolia, China. Eleven quantitative traits related to canopy architecture, yield, and quality were measured during the taproot expansion stage. Root yield was negatively correlated with sugar content. By integrating Spearman correlation analysis with Random Forest (RF) regression, five complementary indicators, including canopy width, SPAD, leaf length, plant height, and lower leaf angle, were retained for multi-trait evaluation. Principal Component Analysis and Ward’s hierarchical clustering further classified the accessions into High-Yield, High-Sugar, and Balanced phenotypic typologies with distinct combinations of yield, sugar content, and canopy traits. A performance-adjusted RF-TOPSIS framework was subsequently used to rank the 216 accessions. BV82 achieved the highest relative closeness coefficient, and all top 20 accessions belonged to the High-Yield group. Sensitivity analyses indicated that the overall ranking was generally stable under alternative weighting schemes and moderate perturbations of criterion weights. These results provide candidate canopy traits and germplasm resources for further field validation and sugar beet breeding.