Genotype-by-Environment Interactions Under Water-Limited Conditions

Abstract

From a plant breeding and statistical perspective, genotype-by-environment interactions (GEI) for traits are considered to have occurred when there is a change in the relative performance of genotypes for the traits between different sets of environmental conditions. When the target population of environments (TPE; Comstock, 1977) for the breeding program consists of environments where water limitation occurs to varying degrees (over locations and/or years), GEI for end-product traits, such as grain yield, greatly complicate the process of selection for broad adaptation to the TPE. Although the presence of GEI may indicate the possibility of selection for specific adaptation, the GEI will need to be understood to enable such targeted selection (Ceccarelli, 1996; Cooper et al., 1996). The incidence of GEI can be visualized in several ways. Some typical examples taken from the results of a wheat (Triticum aestivum L.) multienvironment trial (MET) conducted to test advanced breeding lines under well-watered, drought, and low-nitrogen conditions in northeastern Australia are shown in Figure 3.1 (Cooper et al., 2001). When there are a few genotypes and environments, the trait performance profiles for the individual genotypes can be plotted directly (Figure 3.1a). When there are many genotypes in a few environments, scatter diagrams can be used to compare specific environments (Figure 3.1b). When there are large numbers of genotypes and environments, summaries based on exploratory statistics, such as group means from a cluster analysis (Figure 3.1c) or a biplot (for a single trait) or joint plot (for multiple traits) from an ordination analysis (Figure 3.1d), can be informative.

Topics & Concepts