Analysis of genotype and genotype × environment interaction in durum wheat in warm rainfed areas of Iran

The objective of this investigation was to evaluate the magnitude of G × E interaction effects on durum wheat grain yield and to identify superior genotypes adapted to the test environments. Twenty improved durum wheat genotypes were tested in five locations over three growing seasons. Combined ANOVA indicated that the effect of year (Y) was significant and that of the location (L) was not, but that their interaction (Y × L) was highly significant. The main effect of genotype was also significant, as was the genotype × year interaction (G × Y); genotype × location interaction (GL) was not significant, but three-way interactions (G × Y × L) were highly significant. Clustering of genotypes based on intercept and slope parameters of the linear regression model produced three distinct groups, while using only line slopes for clustering produced no groups at all. The coefficient of determination of the linear regression model ranged from 0.84 (G10) to 0.98 (G2); therefore, it can be concluded that this clustering method was somewhat useful for this data set. According to the dendrogram of clustering based on G and G × E interaction of ANOVA, there were 15 genotypic groups, while according to the dendrogram of clustering based on G × E interaction of ANOVA, there were 12 genotypic groups. Considering all clustering methods and mean grain yield, genotypes G8 (2590 kg ha-1) and G13 (2592 kg ha-1) were superior and thus can be recommended as candidates for release in warm rainfed areas of Iran.