Genotype × Environment Interaction for Leaf Dry Matter Yield in Guineagrass: Identification of Mega‐Environments and Optimization of Multi‐Environment Trials

Guineagrass ( Megathyrsus maximus B.K. Simon & S.W.L. Jacobs [syn. Panicum maximum Jacq.]) presents the highest forage production potential among apomictic cultivated tropical forage grasses in Brazil, and the development of superior cultivars depends on accurate genotype evaluation across multiple environments. This study aimed to characterize genotype × environment interaction (GEI) for leaf dry matter yield (LDMY), identify mega‐environments and environmental covariates associated with GEI, detect genotypes with broad or specific adaptations and define the optimal number of harvests for multi‐environment trials in guineagrass breeding. Twenty‐three genotypes were evaluated in a randomized complete block design across five sites representing three Brazilian biomes, with 11 harvests spanning rainy and dry seasons over 2 years. Data were analysed using mixed models with factor analytic structure for genotypic effects within environments, and different covariance structures for residual effects, with model selection based on the Bayesian information criterion. Genotypic scores and environmental loadings from the selected model were used to identify mega‐environments, assess genotype stability and responsiveness and correlate environmental loadings with climatic covariates. Significant GEI was detected, and three mega‐environments were identified. Minimum temperature was the climatic covariate most strongly associated with the first FA factor underlying GEI. Genotypes differed in adaptability, with both broad and mega‐environment‐specific adaptation detected. Six cumulative harvests were sufficient for reliable genotype selection, and dry‐season harvests showed lower ability to discriminate genotypes for LDMY in guineagrass.