Abstract:
Cowpea [Vigna unguiculata (L.) Walp] is the world’s most important protein source where it is grown for forage, green pods, and grains. Stability performance of cowpea genotypes across contrasting environments is essential for the successful selection of stable and high yielding genotypes. Genotype by environment interaction is one of the leading factors that disturb the stability of genotypes. Therefore, this study was conducted to estimate the effects of genotype, environment, and genotype x environment interaction on grain yield and yield-related traits and to assess the stability of cowpea genotypes for yield across different cowpea growing environments of Ethiopia. Twenty four cowpea landraces and one check were evaluated in 5x5 triple lattice design during 2018/19 cropping season at six environments. Data were collected on yield and yield-related traits: days to flowering, days to maturity, plant height, pods per plant, seeds per pod, grain yield, and hundred seed weight. The analysis of variance for each environment and across environments showed significant differences among genotypes, environments, and GEI for phenological and agronomic traits including yield. From AMMI analysis, environment, genotype, and GEI had 29.79%, 15.6% and 42.06% contribution to the total sum of squares, respectively for grain yield. A large sum of squares for interaction and environment indicated that the environments were diverse, with large differences among the interaction and environmental means causing most of the variation in grain yield. G24 (2632 kg ha-1), G16 (2290 kg ha-1), G2 (2276 kg ha-1), G4 (2250 kg ha-1) and G20 (2213 kg ha-1) were the highest yielder and stable genotypes, respectively with mean grain yields above the grand mean (2049.28 kg ha-1). Simultaneously, G24 and G16 genotypes were the most stable genotypes with mean grain yield above the standard check (2273 kg ha-1). Various stability models: AMMI stability value (ASV), genotype selection index (GSI), cultivar superiority (Pi), Wricke’s ecovalence (Wi), regression coefficient (bi) and devotion from regression (S2di), as well as GGE biplot were used to identify stable genotypes. Accordingly, G24, G16, G4, G20, and G2 were the most stable genotypes with high mean grain yield across all locations. The GGE biplot identified Miesso and Sirinka as more discriminating environments. Miesso was identified as the most representative and an ideal testing environment, which was able to provide unbiased information about the performance of the tested genotypes, whereas Sirinka was identified as the least representative testing environment. Even if the information generated through this research had important implication in GEI study for identifying stable genotypes, further study on more diverse locations and seasons is required to strengthen the result of the current study and to generate more reliable information.