Interação genótipo x ambiente em soja com ênfase na estratificação ambiental para a região central do Brasil

• Autor(a) principal: Branquinho, Rodrigo Gomes
• Data de Publicação: 2011
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• dARK ID: ark:/38995/001300000cpqv
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tde/2937
• topic: zoneamento ambiental; mega-ambiente; VCU; análise AMMI; genótipo vencedor; locais-chave; environment zoning; mega-environment; AMMI analysis; winner genotype; GENETICA::GENETICA VEGETAL

The objective of this study was to establish a consistent environmental stratification for the region of soybean cropping in Central Brazil, based on genotype by environment (GE) interaction analysis. For this, yield data from variety trials conducted by Embrapa Cerrados in partnership with others Brazilian institutions, during seven growing seasons (2002/03 to 2008/09), were used. The study covered six experimental sets that were related to the genotypes of three maturity groups (early, medium and late), and two commercial groups (soybean conventional and transgenic RR), totaling 559 trials analyzed. The statistical treatment of data was performed in two stages: first, analyses of variance were performed for each experiment, from which the estimates of treatment mean (combination of genotype and environment) were obtained. In the second stage the joint and GE interaction analyses were performed. Thus, the yield mean of each genotype in each environment were submitted to the AMMI analysis (Additive Main effects and Multiplicative Interaction model), that led to choose a model with only one principal component (AMMI1). As result of this analysis, the genotypes and environments were jointly represented in a scatter plot called biplot (graph that display the rows and columns of a matrix; in this case, genotypes and environments are marginal in this table). To stratify the target region, the approach of winner genotypes (Gauch & Zobel, 1997; Crop Sci. 37: 311-326) was used. In this approach each stratum is composed by locations that shared a same winner genotype (one that is the higher yielding mean ranking of a location). In the AMMI1 biplot, the boundaries of each stratum were identified by horizontal lines drawn from the ordinate points (scores) corresponding to the environment of transition between two strata, which are characterized by their winner genotypes. With this information, the environmental strata were established for each growing year and experimental set. The maturity groups of assessed lines determined the environmental stratification obtained. Thus, the following locations were grouped to other localities, presenting a characteristic of redundancy: a) early maturity group (seven strata): (Campo Novo do Parecis, Maracajú, São Miguel do Araguaia, Tangará da Serra); (Conquista, Nuporanga, Sidrolândia, Sorriso); (Cristalina, Iraí, Sacramento); (Montividiu, Sonora, Tapurah); (Capinópolis, Senador Canedo); (Guaíra, Morro Agudo); and (Lucas do Rio Verde, Sapezal); b) medium maturity group (four strata): (Anápolis, Montividiu, Tangará da Serra); (Barreiras, Campo Novo do Parecis, Uberaba-Chapadões); (Chapadão do Sul, Conquista, Maracajú, Sonora); and (São Gabriel, Sorriso, Uberaba-Epamig); c) late maturity group (five strata): (Campo Novo do Parecis, Planaltina, Senador Canedo, Tapurah); (Iraí, Sacramento, Sonora); (Lucas do Rio Verde, Sorriso); (Goiatuba, Tangará da Serra); and (Barreiras, São Desidério). Were also identified key-locations to conduct the trials in the final stage of genotypic evaluation (advanced variety trials): a) early maturity group: Anápolis, Barretos, Campos de Júlio, Capinópolis, Chapadão do Céu, Chapadão do Sul, Goiatuba, Igarapava, Jataí, Luziânia, Morro Agudo, Planaltina, Primavera do Leste, Sacramento, São Gabriel do Oeste, São Miguel do Araguaia, Sapezal, Sidrolândia, Sonora, Uberaba-Chapadões, Uberaba-Epamig e Unaí;b) medium maturity group: Barreiras, Barretos, Campo Alegre, Campos de Júlio,Capinópolis, Chapadão do Céu, Chapadão do Sul, Cristalina, Goiatuba, Iraí, Jataí, Lucas do Rio Verde, Luziânia, Montividiu, Perolândia, Planaltina, Primavera do Leste, Rio Verde, Sacramento, São Desidério, Senador Canedo, Sorriso e Unaí; c) late maturity group: Anápolis, Campo Alegre, Campo Novo do Parecis, Campos de Júlio, Capinópolis, Chapadão do Céu, Chapadão do Sul, Cristalina, Goiatuba, Jataí, Luziânia, Montividiu, Primavera do Leste, Rio Verde, São Desidério, São Gabriel do Oeste, Sonora, Sorriso, Uberaba-Chapadões, Uberaba-Epamig e Unaí. Finally, among the locations recommended for the network of advanced trials, one was also appointed as key-location to conduct the initial stages of genotypes assessment in each maturity group. The locations Campos de Júlio (to early group), Rio Verde (medium and late groups) were in order indicated because resulted the best rankings of the winner genotypes through the target region.