Identifying mega-environments to enhance the use of superior rice genotypes in Panama

The objective of this work was to evaluate three methods to identify mega-environments, in order to optimize the use of the genetic potential of rice cultivars during the selection process and to make recommendations for commercial plantations in Panama. Experimental data were obtained from the test performance, between 2006 and 2008, for early maturing cultivars. To achieve the stratification of environments and define mega-environments, the winner genotype method by the AMMI1 model, GGE biplot model and cluster by Ward's method supplemented by GGE biplot were used. The three methods used identified two mega-environments, where the outstanding cultivars were Fedearroz 473 e Idiap 145-05. There was 100% coincidence in the grouping of the cluster x the GGE biplot, with 95.2% coincidence between the AMMI1 x cluster and GGE biplot x AMMI1. The most stable genotype, in both mega-environments, was the Idiap 145-05 cultivar, which indicates its broad and specific adaptive capacity. The adaptive capacity of the superior genotypes and not the agroclimatic conditions of the assessed localities was responsible for defining the mega-environments.