The effect of environmental variables on the genotyping-by-environment interaction in blueberry

Blueberry (Vaccinium corymbosum) has an important role in the global market, with consumption increasing steadily in the past decades. Adapted to multiple environments, a relevant question when investigating fruit quality attributes is the performance of different genotypes over multiple environmental conditions. In this study, we hypothesize that the use of environmental covariables (ECs) can improve the understanding of genotype-by-environment interaction and guide breeders’ decisions. To test it, we used twenty-four genotypes, evaluated across five harvest seasons (2018–2022) in four mega-environments in Florida (Citra FL, Central-North FL, Central FL and South FL), and investigated the impact of twenty-one environmental covariables on the phenotypic expression of five fruit quality traits (weight, total soluble solids, total titratable acidity, firmness, and size). Our contributions in this study are three: (I) first, we identified important temporal windows affecting the phenotypic plasticity for multiple traits; (II) using the environmental covariables, we draw attention on the importance of ECs that characterize the mega-environments and use such information in a mixed model framework to explain the genotype-by-environment interaction; and finally (III) we used linear regression (in the form of Finlay-Wilkinson regression) to estimate adaptability and stability metrics to select promising genotypes. Overall, these findings offer insights into the utility of environmental variables in explaining genotype stability and plasticity, thus, providing a framework to enhance predictive responses and optimize resource allocation in blueberry breeding.