TY - JOUR AU - Acevedo-Barona, Marco Antonio AU - Silva-Diaz, Rubén AU - Álvarez-Parra, Rosa AU - Torres-Angarita, Orlando AU - Reyes-Ramone, Edicta PY - 2019/11/25 Y2 - 2024/03/28 TI - Environmental stratification of rice by genotype x environment interaction analysis using five methods JF - Agronomía Mesoamericana JA - Agron. Mesoam. VL - 31 IS - 1 SE - Articles DO - 10.15517/am.v31i1.35187 UR - https://revistas.ucr.ac.cr/index.php/agromeso/article/view/35187 SP - 43-57 AB - <p><span class="CharOverride-4"><strong>Introduction.</strong> </span>Venezuela has more than two million hectares suitable for rice production, of which about the 25% are exploited, under the production system with flood irrigation with various agronomic management technologies and environments that could affect the phenotypic expression of the genetic materials.<strong><span class="CharOverride-4"> Objective. </span></strong>The objective of this work was to quantify the magnitude of the environmental genotype interaction (IGA) to stratify the irrigated rice environments through five analytical methods and determine their association. <strong><span class="CharOverride-4">Materials and methods.</span></strong> Eight rice genotypes were evaluated in twelve environments resulting from the combination of localities and planting seasons in the regions of the Central and Western Plains of Venezuela, during the years 2012-2013. The stratification of environments was made based on five methods: traditional Lin (TL), Euclidean distance (ED), simple percentage of IGA (% SP), Pearson correlation (r<span class="CharOverride-5">xy</span>), and factor analysis (AF). The randomized complete block statistical design was used with three repetitions in 20 m<span class="CharOverride-3">2</span> plots. <strong><span class="CharOverride-4">Results.</span></strong> ANOVA detected significant IGA, explaining 35% of the total variation. Favorable environments for rice represented 33%. The methods used TL, ED, % SP, rxy, and AF grouped the twelve irrigated rice environments into 10, 5, 8, 10, and 3 groups respectively; they were not efficient in identifying different environments when different planting seasons were used in the same locality. <span class="CharOverride-4"><strong>Conclusions.</strong> </span>The factor analysis method was more efficient in identifying homogeneous environments, complemented with % SP and TL methods that presented moderate association. The Bancos of San Pedro and Asoportuguesa environments were more informative and indicated for the evaluation of rice genotypes. The opposite occurred with the Araure, Algodonal, and Torunos localities.</p> ER -