Genotype-by-Environment Interaction and Stability Analysis for Grain Yield in Durum Wheat Using GGE Biplot and Genotypic and Environmental Covariates

Abstract

Understanding the genetic and environmental causes of genotype-by-environment (GE) interaction for grain yield is of fundamental importance in plant breeding. This study aimed at investigating the GE interaction and the stability of durum wheat (Tritium turgidum L. ssp. durum) genotypes evaluated for grain yield across different locations and years, using empirical and analytical models. The study used 19 genotypes in 14 environments, representative of rainfed durum wheat-growing areas. The genotype main effect plus GE interaction (GGE) biplot model, partial least squares regression and factorial regression models were applied for data analysis. The combined ANOVA revealed significant genotype, environment and GE interaction effects, with the environmental main effect as a main source of variation (77.9% of total variation). The mean yield of the genotypes ranged from 486 to 5594 kg/ha across environments. Using GGE biplot analysis, the test environments were classified into four groups, each with different winning genotypes. Based on mean yield and stability performance across environments, breeding lines G17 and G15 significantly out-yielded the best national check and could be recommended for release as new varieties. Rainfalls in March, June, November and May, average temperatures in June and May, heading date and 1000-kernel weight were among the explanatory covariates that significantly (P < 0.01) affected the GE interaction for grain yield.