AMMI and GGE Models Analysis of Stability and GEI of Common Bean (Phaseolus vulgaris L.)Lines in Ethiopia

Abstract

When genotypes were introduced into a new and diverse production environments, occurrence of significant GEI (genotype by environment interaction) complicates selection of stable genotypes. Therefore, fifteen small red common bean lines introduced from CIAT including one standard check (Nasir) were evaluated at five representative dry bean growing locations of Ethiopia for seed yield performance using 4x4 triple lattice design in the 2013 and 2014 main cropping seasons to estimate the magnitude of GEI effects and to identify broadly or specifically adapted lines.  Combined ANOVA, AMMI and GGE biplot models were used to interpret the data.  Both the main and interaction effects were highly significant (P < 0.01) and  Environment, Line, and GEI explained 81.91%, 15.5% and 2.2% of variation in treatment structure, respectively,  and indicated greater influence of location and importance of simultaneous consideration of mean performance and stability. IPCA₁ and IPCA₂ were highly significant (p < 0.01) and together contributed nearly 60% variation in the GEI sum of squares. AMMI 1, GGE ranking, and GGE comparison biplots enabled identification of broadly adapted lines. Lines, KG-71-1, KG-71-23, and KG-71-44 were selected  as broadly adapted lines. GGE biplot analysis suggested presence of four mega-environments and enabled identification of specifically adapted lines. However, the specific adaptability of lines was not repeated over years and thus, GEI couldn't be exploited and broadly adapted lines  were recommended for verification and release.

Keywords: AMMI, GGE, Adaptability, GEI, Broad adaptation, Line, Environment