Heterogeneous Variances in Multi-Environment Yield Trials for Corn Hybrids

Recent developments in statistics and computing have enabled much greater levels of complexity in statistical models of multi-environment yield trial data. One particular feature of interest to breeders is simultaneously modeling heterogeneity of variances among environments and hybrids. Our objective was to estimate the level of heterogeneity of genotype by environment interaction variance and error variance in the Iowa Crop Performance Test for Corn. A Bayesian approach was used to estimate variance components in a hierarchical model that allows for heterogeneous error and genotype-by-environment interaction (GEI) variances applied to corn yield data from the Iowa Crop Performance Test performed between 1995 and 2005. An average of 508 hybrids were tested per year with very little overlap between locations and years, which resulted in a very unbalanced data set. We divided the data into 16 subsets to study the effect of variability across locations and years. We found GEI and error variances to be heterogeneous among both environments and genotypes. Our results for corn contrasted previous work on oat (Avena sativa L.) in which very little heterogeneity was found for error variance among cultivars suggesting that different corn (Zea mays L.) hybrids can have different genotype by environment interaction variances and different error variances.