Evaluation of Yield Stability in Finger Millet (Eleusine coracana L.) Genotypes Using Multivariate Approaches

Finger millet is commonly cultivated in the semiarid tropics, where it is primarily grown by subsistence farmers. However, grain yield remains low due to the complex quantitative nature of the trait and its low heritability. Therefore, genotype × environment interaction (GEI) significantly influences yield production. This study investigates the impact of GEI on the performance of finger millet genotypes across multiple environments, emphasizing the crop's sensitivity to climate variability. The objectives of this study were to evaluate the effects of genotype, environment, and GEI on yield and identifying high-yielding stable genotypes. Multienvironment trials (METs) were conducted at Axum, Negele Arsi, and Assosa during the 2018 and 2019 in summer cropping seasons utilizing row-column designs and advanced statistical analyses, including additive main effects and multiplicative interaction (AMMI) and genotype and genotype × environment interaction (GGE) biplot analyses. AMMI analysis indicated substantial environmental effects, with interaction principal component axes accounting for over 80% of the GEI. The GGE biplot identified the relationships between environments, highlighting specific genotypes that are optimal for each environment. Genotype G_32, with a yield of 2.75-ton ha^-1, showed the highest mean yield values and the highest stability metrics using mean ranks and cultivar superiority stability values of 0.12 and of 9.0, respectively. Genotype G₅₃ was the most stable, with a variance of ranks of 17.60, mean absolute difference of pairs of ranks of 4.90, and Wricke's ecovalence of 0.02. The choice of stability measures is critical, depending on plant breeders' objectives and the heritable traits targeted. Hence, genotype G₃₂ had the highest grain yield performance and the most stable genotype and recommended for wider production in finger millet growing areas. Finally, the study demonstrates that AMMI and GGE are effective methods for selecting superior genotypes in diverse environments, providing valuable insights for finger millet breeding programs.