GENOTYPE X ENVIRONMENT INTERACTION AND STABILITY ANALYSIS FOR SUGARCANE GENOTYPES EVALUATED IN MULTI-LOCATION TRIALS

Selection for wider adaptability of sugarcane varieties to specific production environments is often challenged by the occurrence of significant genotype by environment interactions (GEI). Several statistical models have been proposed for exploiting positive GEI and supporting decisions in varietal selection and recommendation for target environments. Additive main effects and multiplicative interactions (AMMI) model effectively captures the additive and multiplicative components of GEI and provides meaningful interpretation of multi-environment data sets in breeding programs. The objective of this study was to assess the significance and magnitude of GEI effects on cane yield and suggest the exploitation of the positive GEI effects using AMMI biplot analysis. Cane yield data of nine sugarcane genotypes evaluated at seven locations for two crop seasons (2008-09 and 2009-10) across north-west zone of sub tropical India were used for this purpose. AMMI analysis of variance of cane yield (t ha-1) showed that 61.11% of the total sum of squares was attributable to GEI effects indicating high interaction among genotypes and environments. However, 22.34% and 16.05% of the total sum of squares were attributable to environments and genotypes, respectively. The sum of squares of interaction principal component analysis axis –I (IPCA-I) and IPCA-II were significant and cumulatively contributed to 69.40% at 38 df. According to the AMMI biplots, none of the varieties had superior performance in all the environments. However, genotype CoS 767, combining low absolute IPCA-I score with good cane yield, was the overall winner with less variable yields across the environments which was also confirmed by low estimate of AMMI stability value.