Analysis and Interpretation of Sugarcane Cultivar Adaptability to Time of Harvest in South Africa.

The adaptability of new sugarcane (Saccharum spp.) cultivars to time of harvest (TOH), and reasons for yield variability between harvest times are currently unclear in South Africa. The objectives of this study were to evaluate the adaptability of commercial cultivars to TOH and to identify environmental factors responsible for differential cultivar responses.  Two field trials consisting of the same set of ten commercial cultivars were established adjacent to each other and harvested either early (April/May) or late (October/November) in the season for six crops. Climatic data were summarized within individual growth stages (establishment, elongation, ripening) and used to create environmental covariates for temperature (TT), rainfall (RAIN), radiation (RAD), and a water stress index (WSI). Additive main effects and multiplicative interaction (AMMI) analysis, principal components analysis (PCA), factorial regression, and biplot analysis was used to analyse and interpret the genotype x environment (G x E) interaction for three commercial variables.  The G x E interactions were significant (P<0.001) for all three variables. The AMMI biplots showed distinct separations between environments, and characterized cultivars according to their adaptability to early (NCo376, N36, N35, N29) and late (N27, N19, N17) harvesting. PCA and factorial regression identified RAD and TT as significant environmental covariates affecting TOH, while WSI and RAIN had negligible effects. The covariate-effect biplot identified correlations between plant population and early harvests, while stalk mass was correlated to late harvests. The implications of these findings are discussed in relation to selection strategies and future interpretive studies involving sugarcane G x E interactions.