Integrating qualitative and quantitative traits with multivariate analysis to decipher the diversity structure of barley landraces for sustainable agriculture
Understanding genetic diversity in barley (Hordeum vulgare L.) landraces is crucial for identifying potential parental lines and conserving valuable genetic resources. This study evaluated the agro-morphological and genetic diversity of seventy-five barley landraces under the field of College of Natural Resource Management (CNRM) Bardibas of Madhesh Province, Nepal, to identify valuable germplasm for breeding and conservation for Terai condition. The experiment was conducted in Alpha Lattice design with two replications. Qualitative and quantitative data were collected at different stage of barley. Multivariate test and Analysis of variance (ANOVA) test were conducted to test the diversity of barley. The ANOVA revealed highly significant genotypic differences (p < 0.001) for key agro-morphological traits, including days to flowering (78.5–123 days), days to maturity (115–148 days), spike length (4.4–19.4 cm), filled grains per panicle (7.4–59.8), sterility (5.6–84.6%), plot yield (3.0–181.2 g plot− 1), and plot biomass (25–475 g) confirming a broad phenotypic basis for selection. Substantial phenotypic diversity was observed with Shannon-wiener diversity indices reaching up to 0.99 for grain-related traits such as grain surface (0.994), rachilla hair (0.997), grain crease width (0.976), and spike density (0.899). Principal Component Analysis (PCA) revealed that the first two principal components (PC1 and PC2) explained 54.89% of the total variance (PC1 = 42.63%, PC2 = 12.26%). Traits with the highest loadings on PC1 included days to maturity (0.931), sterility (0.865), spike length (0.865), and days to flowering (0.889), while PC2 was strongly influenced by thousand-grain weight (0.739). Cluster analysis using both K-means (K = 2) and hierarchical Ward’s methods consistently grouped the landraces into four distinct clusters, with a high degree of concordance between the two methods. The PCA bi-plot and heat-map visualizations clearly differentiated these clusters and highlighted specific landraces (NGRC7732 and NGRCO7726) occupying extreme positions, indicating unique genetic backgrounds. The results highlight the presence of valuable genetic variation and adaptive traits for suitable for barley improvement ad breeding programs aimed at enhancing yield stability and resilience.