Jump to Main Content
Integrating different stability models to investigate genotype × environment interactions and identify stable and high-yielding barley genotypes
- Vaezi, Behrouz, Pour-Aboughadareh, Alireza, Mohammadi, Rahmatolah, Mehraban, Asghar, Hossein-Pour, Tahmasb, Koohkan, Ehsan, Ghasemi, Soraya, Moradkhani, Hoda, Siddique, Kadambot H. M.
- Euphytica 2019 v.215 no.4 pp. 63
- analysis of variance, barley, genotype, genotype-environment interaction, grain yield, growing season, models, plant breeding, principal component analysis, Iran
- Barley is the fourth largest grain crop globally with varieties suited to temperate, subarctic, and subtropical areas. The identification and subsequent selection of superior varieties are complicated by genotype-by-environment interactions. The main objective of this study was to use parametric and non-parametric stability measures along with a GGE biplot model to identify high-yielding stable barley genotypes in Iran. Eighteen barley genotypes (16 new genotypes and two control varieties) were evaluated in a randomized complete block design with four replications at five locations over three growing seasons (2013–2014, 2014–2015, 2015–2016). The combined analysis of variance indicated that the environment main effect accounted for > 69% of all variation, compared with < 31% for the combined genotype (G) and genotype-by-environment interaction effects. The mean grain yield of each genotype across the five test sites and three seasons ranged from 1900 to 2302 kg ha⁻¹. Using Spearman’s rank correlation and principal component analyses, the stability measures were divided into three groups: the first included mean yield, TOP and b, which are related to the dynamic concept of stability, the second comprised θᵢ, W ᵢ² , σ ᵢ² , CVi, [Formula: see text], KR, and the non-parametric measures, S⁽ⁱ⁾ and NP⁽ⁱ⁾, which are related to the static concept of stability, and the third included θᵢ and R². The GGE biplot analysis indicated that, of the five test locations, Gonbad and Moghan had the most discriminating and representative environments. Hence, these locations are recommended as ideal test locations in Iran for the selection of superior genotypes. The numerical and graphical methods both produced similar results, identifying genotypes G12, G13, and G17 as the best material for rainfed conditions in Iran; these genotypes should be promoted for commercial production.