Grain yield stability and genotype by environment interaction of quality protein maize hybrids in Ethiopia

Beyene, D., Wegary, D., Tesso, B., Jalata, Z., Geleta, N., & Sadessa, K. (2025). Grain yield stability and genotype by environment interaction of quality protein maize hybrids in Ethiopia. Agrosystems Geosciences & Environment, 8(4), e70236. https://doi.org/10.1002/agg2.70236

Maize (Zea mays L.) production in Ethiopia spans across various agro-ecologies, encompassing humid highland, humid midland, dry lowland, and humid lowland areas. Identifying well-adapted and productive genotypes for target production environments could be achieved by evaluating new experimental hybrids across various representative test environments. This study aimed to examine the mean grain yield performance, grain yield stability, and genotype-by-environment interaction of quality protein maize (QPM) hybrids evaluated across environments in Ethiopia. Forty-eight QPM experimental hybrids, along with two commercial check hybrids, were evaluated across six environments. Analysis of variance for grain yield exhibited highly significant (p ≤ 0.001) differences due to genotype, environment, and genotype by environment interaction (GEI). Additive main effect and multiplicative interaction (AMMI) analysis revealed that genotype, environment, and GEI effects contributed to 4.57%, 78.59%, and 16.84% of the total variation, respectively. The first two interaction principal component axes (IPCAs) explained 66.29% of the total variations attributed to GEI sum of squares, indicating that these IPCAs captured most of the interaction effects. The AMMI stability value identified G5, G19, G22, and G42 as stable and high-yielding QPM hybrids, while G5 was the most stable genotype identified by yield stability index analysis. Genotype main effects plus GEI (GGE) biplot analysis identified G13, G14, and G25 as the most desirable hybrids. Among the test environments, Holeta was identified as an ideal test environment, exhibiting the highest discriminating power among the tested hybrids and the most representative of the test environments. The polygon view of GGE biplot subdivided the testing environments into different groups, mainly represented by Holeta, and Haramayaand Kulumsa. Among the various analytical models, the GGE biplot proved to be the most effective and precise tool for identifying high-yielding and stable hybrids. Results of this study indicated the possibility of developing stable and high-yielding QPM hybrids suited to representative maize production environments.