An Assessment of Taro Yield and Stability Using Ammi and GGE Biplot Models

The yield stability of eight taro (Colocasia esculenta (L) Schott) genotypes across two locations in two years was assessed using the Additive Main Effect and Multiplicative Interaction (AMMI) and Genotype and Genotype-by-Environment (GGE) biplot models. The results of combined analysis of variance for the yield of the taro genotypes grown in 4 environments showed that yield was significantly affected by environments (E), genotypes (G) and genotype by environment interactions (GEI). Differences between genotypes and environments accounted for 24.13% and 56.41% of the total variation respectively while genotype x environment interaction accounted for 9.03% of the total variation. The first interaction principal component axis (IPCA) from the AMMI analysis accounted for 71.10% of variation due to GEI. The biplot accounted for 97.09% of the treatment sum of squares. Both AMMI and GGE models identified NCe 005, NCe 011 and NCe 010 as most stable, but NCe 010 with the highest yield was rated the best genotype across the environments. As a result of the study, E3 was selected as the favourable test environment for the taro yield multi-environment trial. The result showed that application of AMMI and GGE biplots facilitate visual comparison and identification of superior genotype for each target environment.