Study on Correlation and Repeatability of Parametric and Multivariate Statistics of Grain Yield Stability in Rainfed Barley

Multi-environment trial data are required to obtain stability performance parameters as selection tools for effective cultivar evaluation. The interrelationship among several stability parameters and their associations with mean yield, along with the repeatability of these parameters in consecutive years was the objective of this study. Barley yield data of 18 cultivars, proprietary of Dryland Agricultural Research Institute, evaluated in 12 locations over 2005-2007 in three locations of Iran was used for the combined analysis of variance in three datasets. I: Across locations in a single evaluation year (dataset A), II: Across locations in each of two single evaluation year (dataset B), III: Across all locations in three years (dataset C). For each dataset, cultivar phenotypic variance and other statistics were appropriately partitioned in its components. The interrelationship among the parameters and their associations with mean yield based on Spearman rank correlation was studied in each of the three single evaluation years (dataset A). Rank correlation coefficients were also used as estimates of the repeatability of these stability parameters across two year combinations (dataset B). The parameters 2 σi Shukla and 2 Wi were consistently highly correlated with each other but not with mean yield in all single and pair evaluation years. The parameters bi , 2 Si and CV were consistently highly correlated with each other but not with mean yield in all single and pair evaluation years. Result showed that Shukla variance, Wricke ecovalence, SIPC1 and ASV AMMI stability statistics have high repeatability. Thus it is suggested that more stability statistics especially nonparametric methods be used for determination of repeatability methods.