查看文章

Matching soybean variety selection with its production environment is often challenged by the occurrence of significant genotype-by-environment interactions (GEI) in the variety development process. Several statistical models have been proposed for increasing the chance of exploiting positive GEI and supporting breeding program decisions in variety selection and recommendation for target set of environments. Additive main effects and multiplicative interactions (AMMI) and site regression (SREG) genotype plus genotype-by-environment interaction (GGE) models are among the models that effectively capture the additive (linear) and multiplicative (bilinear) components of GEI and provide meaningful interpretation of multi-environment data set in breeding programs. The objective of this study was to assess the significance and magnitude of GEI effect on soybean grain yield and exploit the positive GEI effect using AMMI and SREG GGE biplot analysis. Grain yield data of 11 genotypes evaluated at 4 sites for three cropping seasons (2002, 2003 and 2004) across the soybean production ecology in Ethiopia were used for this purpose. AMMI analysis showed that grain yield variation due to environments, genotypes and GEI were highly signifiscant (p< 0.01). Environments explained the greater proportion (61.08%) of total yield variation followed by GEI (34.13%) and genotypes (4.79%), indicating the necessity for testing soybean varieties at multi-locations and over years. The first five bilinear AMMI model terms were highly significant (p< 0.01) and of which the first two terms explained 67.5% of the GEI. According to the AMMI and SREG GGE …