Controle de qualidade e interação genótipo ambiente em ensaios de competição de genótipos de arroz

Abstract

The objectives of this study were to estimate the limits for classifying the experimental precision, to verify significance of genotype vs. environment interaction and to identify the adaptability and stability of rice genotypes among groups of cultivation environments. Grain yield data of 24 yield trials coming from crop seasons 2016/17 and 2017/18 in eleven municipalities of Rio Grande do Sul and two from Santa Catarina. The experiment design was randomized blocks with 3 replications and 9 genotypes. There were determined the limits of class for experimental precision using pseudo-sigma of the coefficients of variation and the least significant difference. Heritability and selective accuracy were estimated. Combined analysis was done by crop season. Environment stratification and decomposition of the interaction were divided in simple and complex. Adaptability and stability study was done based on the methods of linear and simple regression and non-parametric analysis. The quality of the inferences taking into consideration the answers obtained by the very high selective accuracy (>0,90) to high (>0,70), general coefficient of variation, intermediate (≤10,86%) to low (≤1,62%) and general least significant difference, from intermediate (≤31,53%) to low (≤4,70%) for these trials are of good precision for the variable yield. The environment influenced on the genotype performance for this variable. The cultivars hybrids are superior compared to the conventional ones for the variable yield.