Abstract

The study modeled the yield stability of selected early- and medium-duration cowpea genotypes across six locations in northern Ghana to identify the most stable and winning genotype(s). The models employed in the study included a combined analysis of variance (ANOVA), the Additive Main Effects and Multiplicative Interaction (AMMI) model, the Genotype + Genotype-by-Environment (GGE) biplot model, and non-parametric statistical models. The yield stability and adaptability measure SARI-2-50-80 indicated that the genotypes SARI-2-50-80, IT86D-610, IT10K-837-1, and SARI-6-2-6 are high-yielding and stable among medium-duration cowpea lines across environments. The study also found that genotypes SARI-3-11-80 (0.227 t/ha) and the checker cultivar KIRKHOUSE-BENGA (0.145t/ha) had relatively lower ASV (AMMI Stability Value) values, indicating higher stability in medium-duration cowpea lines. Genotypes SARI-3-11-80 (0.0696t/ha) and SARI-1-3-90 (0.0952t/ha), SARI-13-17-2 and IT07K-299-6IT07K-299-6, demonstrated lower MASI (Mean Absolute Standard Index) values, suggesting higher stability and adaptability. Genotypes SARI-3-11-80 (0.442t/ha) and KIRKHOUSE-B (0.581 t/ha) exhibited relatively lower MASV (Mean Absolute Standard Values /Modified AMMI stability Value)) values, indicating higher stability and adaptability. Damongo, Yendi, and Manga were identified as the most promising environments for most of the lines. Yield stability was significantly predicted by genotype, environment and genotype by environment interactive effect. The ANOVA for confirmation, AMMI and GGE for decomposition and visualization, and non-parametric models for robust validation represent a highly accurate and comprehensive approach to MET (Multi-Environment Trial) analysis. The AMMI model derived the calculated stability metrics which were used for the accurate identification of the high-yielding and stable genotypes.

This work is licensed under a Creative Commons Attribution 4.0 License.