Abstract
Genotype by environment interaction (GEI) markedly influences the success of breeding strategies in a versatile crop such as cowpea (Vigna unguiculata (L.) Walp.). Twenty cowpea genotypes were tested in a randomized complete block design with three replications at Gofa, Kucha, and Humbo in Meher seasons of 2016 and 2017 (E1 to E6) and Belg seasons of 2017 and 2018 (E7 to E12) to quantify and evaluate the effects of genotypes, environments and their interactions for grain yield of cowpea genotypes and to identify stable and/or high-yielding genotypes. The environment, genotype, and GEI effects were highly significant (p < 0.001), with the contribution of 42.3%, 23.0%, and 34.7%, respectively to the TSS. Additive main effect and multiplicative interaction (AMMI), genotype main effects plus genotype-environment interaction (GGE), ASV (AMMI stability value), and Genotype stability index (GSI) were used to identify stable genotypes. The GGE-biplot model showed that the twelve environments used for the study clustered under three mega-environments. Our results showed that IT96D-604(G12), IT-89KD (G16), IT93K-293-2-2 (G14), 93K-619-1(G13), IT97K-569-9(G20), and IT99K-1060(G15) scored the highest grain yield (1.67, 1.62, 1.55, 1.51, 1.51, and 1.45 t ha(-1)), respectively, over environments. AMMI and GGE biplots analyses identified G16 (IT-89KD) and G14 (IT93K-293-2-2) as stable and high-yielding genotypes across environments and can be further tested in variety verification and later on released as varieties and can also be used for different breeding purposes in all cowpea growing areas in southern Ethiopia. The four high-yielding genotypes IT96D-604, 93K-619-1, IT97K-569-9, and IT99K-1060 could be recommended to be included in breeding or variety verification trials for release. Moreover, our results denoted the effectiveness of AMMI and GGE biplot techniques for selecting stable genotypes, high yielding, and responsive.