Multivariate analysis of yield and phenotypic diversity in rainfed rice (Oryza sativa L.) landraces from Gokuleshwor, Baitadi, Nepal

Rice is the predominant food crop in Nepal, contributing to the national food security and the rural economy. The objective of this study was to assess the phenotypic diversity of Nepalese rice landraces under rainfed conditions to identify promising genotypes for future varietal improvement. A total of 22 rice landraces, including one local check (Rato Basmati), were evaluated during the 2024 rainy season at Gokuleshwor, Baitadi, Nepal, using an alpha lattice design with two replications. Highly significant differences were observed for plant height, days to flowering, days to maturity, panicle length, and thousand grain weight. Non-significant variation was noted for grains per panicle, anthesis-flowering interval, grain sterility percentage, biological yield, harvest index, and grain yield. Grain yield ranged from 0.605 tons/ha (Saal Dhan) to 6.065 tons/ha (Ajaymeru-1). Grain yield showed a highly significant positive correlation with biological yield (r = 0.81***). Days of maturity showed a highly significant positive correlation with days of flowering (r = 0.91***). Cluster analysis revealed 22 genotypes in four distinct clusters. Principal component analysis identified two major components: PC1 explained 33.3% of the variation, while PC2 accounted for 20.23%. In the mid-hills of Nepal’s Sudurpaschim region, this study successfully identified genotypes of rice that are suitable for rainfed and direct seeding conditions. This study provides important information for future breeding programs that aim to improve rice genotype adapted to rainfed environments with superior agronomic traits to improve food security and assist vulnerable farming communities.