Can spring wheat-growing megaenvironments in the northern Great Plains be dissected for representative locations or niche-adapted genotypes?

Navabi, A., Yang, R.-C., Helm, J., Spaner, D.M. Can spring wheat-growing megaenvironments in the northern Great Plains be dissected for representative locations or niche-adapted genotypes? Crop Science Volume 46, Issue 3, May 2006, Pages 1107-1116

Abstract- Characterizing variety testing sites and identification of sites with negligible genotype × environment crossover interaction is important for plant breeders wishing to identify superior germplasm and (or) cultivars for a wide range of environments. Long-term multilocation grain yield data from the regional hard red spring wheat (Triticum aestivum L.) variety trials from 1981 to 2002 (472 location years assessing 64 wheat genotypes) in Alberta, Canada, were employed for this study. The shifted multiplicative model (SHMM) and the site regression model (SREG) were used to group testing sites into subsets with reduced crossover interaction. Both models identified yearly subsets of testing sites with negligible crossover interaction. However, the yearly site groupings did not generally follow a repeatable pattern over years. Clustering did not correspond with provincial agroclimatic classification, nor did it correspond with site-specific yield potential. Genotype × environment patterns were therefore inconsistent over the years, mainly because of complex, highly variable, and unpredictable year × location effects. We identified sites appearing to be more discriminative and predictive of average genotype performance. This suggests that regional variety trials may be conducted at a fewer more representative locations predictive of average varietal performance. We conclude that the spring wheat growing areas in Alberta (and in the northern Great Plains in general) belong to a single megaenvironment with unpredictable crossover interaction patterns. Because of the highly variable and unpredictable genotype × environment interaction patterns in Alberta, genotypic selection targeting wide adaptation is recommended. Although genotype × environment patterns were not repeatable, the yearly high yielding and stable varieties were repeatedly selected over years. These varieties were the most popular varieties grown by farmers during the testing time period.