Genotype by Environment Interaction and Stability Analysis for Grain Yield of Durum Wheat Genotypes Evaluated in Ethiopia

Abstract

The objective of this study was to examine the genotype by environment interaction and evaluate the stability of genotypes using Additive Main effect  and Multiplicative Interaction (AMMI) and genotype and genotype by environment (GGE) biplot analyses. Accordingly, twenty genotypes were evaluated  across seven locations using RCBD with four replications during 2011/12 main cropping season. Stability of genotypes was calculated using cultivar  superiority measure, Wricke’s ecovalence and AMMI stability values. AMMI analysis showed significant (P<0.001) effects of genotype, location and  genotype by location interaction for  grain yield. Location, genotype and genotype by location interaction accounted 74%, 2% and 7% of the variation in  grain yield, respectively. The mean squares of the first and second interaction principal component axis were highly significant at P≤0.001 & P≤0.01,  respectively. Hence, AMMI model with the first two IPCA was the best predictive model. The stability analysis revealed that genotype 7 (G7) was highly  stable with above average grain yield and can be a promising candidate for release. The GGE biplot analyses revealed that the first and second principal  component explained 46% and 28% of the GGE sum square, respectively. On the ‘which won where pattern’ of the biplot, the test locations were grouped  in to two mega environments. Mega environment-1 included four of the seven locations and represented by the best performer genotype G13. Whereas,  mega environment-2 included two locations and represented genotype G1. Among the test locations, Angacha and Arsi Robe were the most  discriminating locations while Hossana was the least and should not be included as a test location.