Genetic Diversity and Trait-Based Selection in Aromatic Rice (Oryza sativa L.): A Multivariate Trait Analysis Approach

Abstract

Aromatic rice holds substantial value in global and domestic markets due to its distinctive aroma, grain quality, and culinary characteristics. This study evaluated 50 aromatic rice genotypes, including breeding lines and landraces, over two consecutiveKharif seasons (2023–2024) at ANDUAT, Ayodhya, using a randomized complete block design with three replications. The objective was to assess genetic variability, identify key yield-contributing traits, and develop predictive models for grain yield. Thirteen quantitative traits were recorded following the Standard Evaluation System (SES) for rice. Descriptive statistics revealed considerable variability across genotypes, with grain yield ranging from 12.61 g to 62.38 g/plant and biological yield from 49.00 g to 154.00 g/plant. High genotypic coefficient of variation (GCV) and heritability values were observed for yield (256.71%), panicle-bearing tillers (81.80%), and biological yield (42.55%), indicating strong genetic control and potential for selection. Correlation and path coefficient analyses identified plant height (direct effect = 0.295) and tillers per plant (direct effect = 0.215) as critical determinants of yield. Principal Component Analysis (PCA) explained trait-driven variance, while hierarchical clustering grouped genotypes into distinct clusters for diversity assessment. Notably, genotypes AR-6, AR-24, AR-12, and AR-4 outperformed the check variety PB-1121 in multiple traits. With R² values demonstrating strong model fit. The integration of classical statistical tools and modern analytics provides a robust framework for identifying elite genotypes for aromatic rice improvement.