Spatial and seasonal effects confounding interpretation of sunflower yields in Argentina

Sunflower in Argentina is grown in subtropical (northern) and temperate (central and southern) environments. These environments can substantially complicate hybrid selection due to large genotype x environment (G x E) interactions. The spatial and seasonal differences between and within central and northern regions were investigated through two complementary approaches: (i) pattern analysis was applied to a long-term environmental data record, in order to group environments based on the mean value of each variable for the different crop stages in a range of years and locations, (ii) a trial dataset, comprising 68 trials, covering 8 years (Y) and 18 locations (L), was analyzed to quantify G, E, and G x E variance components for grain yield, oil concentration and oil yield. Pattern analysis of environmental data separates northern and central environments over both years and locations. Northern environments had shorter photoperiods, less rainfall and lower temperatures during vegetative phases; shorter photopetiods during flowering, and higher temperatures during grain filling. Seasonal differences in trial yield were highly associated with rainfall during reproductive phases. Residual maximum likelihood analyses detected that 86% (central region) and 83% (northern region) of the total G x E interaction for oil yield was attributable to G x L x Y. The relatively low G x L interaction compared with G x L x Y implies that broad adaptation within regions is required. To achieve repeatabilities of more than 80%, trials would need to be conducted over at least 6 years with 15 locations per year. Environmental differences between regions showed a strong degree of repeatability, and were consistent with differences in magnitude of G x E interactions for oil yield components. (C) 2002 Elsevier Science B.V. All rights reserved.