Genotype-by-environment interaction analysis for grain yield of barley genotypes under dryland conditions and the role of monthly rainfall

In rainfed areas, due to the many variable environmental factors, the introduced cultivars, in addition to high yield, must also have yield stability. In the present study, 15 barley lines along with three local control cultivars under dryland conditions in six cold locations of Iran during three consecutive years (2013–2016) were evaluated. The results showed that the genotype-by-environment interaction is highly significant. The ratios between the sums of squares of G, GE and IPCA1 indicated that the AMMI model is suitable for data analysis. Also, combining AMMI model with BLUP technique and using WAASB index was promising and the results of both were in agreement. With the AMMI1 model, three mega-environments were determined, and the locations of Maragheh, Qamlu and Zanjan were located in the same mega-environment every three years. Finally, it was concluded that the second control cultivar was the most desirable genotype for most environments. Genotypes 16, 15 and 12 with high yield and optimal stability have the ability to be used as new potential genetic resources for improving and stabilizing grain yield in barley breeding programs for cold rainfed conditions. Genotypes 4, 5, and 6 should only be used in narrow adaptation to restricted environments. On the other hand, rainfall in November and January is important for proper seedling establishment, and rainfall in April and May is important for pollination and grain filling period.