Stable yield performance of genotypes is very important in countries like Ethiopia where means to modify environments are limited. However, occurrence of significant genotype X environment interaction (GEI ) complicates selection of stable genotypes. In Ethiopia, the yield potential of common bean varieties is underutilized due to poor addressing of all potential areas and mismatch between selection and production environments. Thus, 14 common bean varieties were evaluated at seven locations for seed yield performance using Randomized complete block design with three replications in the 2010 main cropping season to estimate the magnitude of GLI effects and identify broadly or specifically adapted varieties. Combined ANOVA, AMMI and GGE biplot models were used to analyze data. Both main and interaction effects were highly significant (P<0.01) and location, variety, and GLI explained 50.3%, 28.8% and 20.9% of variation in treatment structure. This indicated greater influence of location and importance of simultaneous consideration of mean performance and stability. PC1 and PC2 were highly significant (p < 0.01) and together contributed more than 79% variation in the GLI sum of squares. Varieties, Zebra-90, Goberasha, Roba-1, Nasir, and Awash-melka were selected as broadly adapted cultivars. GGE biplot analysis suggested presence of two mega-locations and enabled identification of specifically adapted varieties. However, GEI couldn't be exploited from one season trial and farmers in the Zone should grow widely adapted varieties.
Keywords: AMMI, GGE, Yield stability, GLI, Broad adaptation, Genotype, Environment.