Classificação de cultivares de soja adotando abordagem univariada e multivariada

• Autor(a) principal: Carvalho, João Paulo Santos
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFLA
• Texto Completo: http://repositorio.ufla.br/jspui/handle/1/13269
• topic: Interação genótipo por ambientes; Adaptabilidade e estabilidade; Soja - Cultivo; Soja - Índice de seleção; Genotype interaction by environments; Coincidence index; Adaptability and stability; Interação genótipo por ambientes; Soybean - Cultivation; Soybean - Selection index; Fitotecnia

The recommendation of soybean cultivars adapted to region that it will be cultivated is a prime factor of the success of the crop. In this context, the use of selection indices consists in a good alternative considering several traits, simultaneously, throughout the evaluations, allowing the identification of superior cultivars that associate traits of interest. The adaptability/stability methods allow knowing the contribution to Genotype-Environment interaction (G x E) and the risk associated to each cultivar. Therefore, the objective of this study was to identify commercial soybean cultivars that associate early-maturity, good yield performance and be adapted to the southern region of Minas Gerais, adopting univariate and multivariate strategy; to study the G x E interaction considering all cultivars simultaneously, as well as by maturity groups. The experiments were carried out in the crop year of 2014/2015 and 2015/2016 in the state of Minas Gerais, using a randomized complete block design with 35 commercial cultivars, and evaluated in 7 environments, considering locals and crop years combinations. In the univariate analysis were evaluated the following traits: harvest index, number of pods, number of grains, number of grains per pod, yield, weight of 100 grains, daily increment, lodging, plant height, 1 st pod insertion and absolute maturity. The selection indexes were calculated considering the yield, harvest index, plant height, 1 st pod insertion and absolute maturity traits. The coincidence between univariate analysis and selection indexes were performed by using the coincidence index, proposed by Hamblin and Zimmermann (1986). The analysis of each cultivar contribution to G x E interaction and the associated risk were obtained by ecovalence and confidence index methods respectively. The results allowed inferring NS 7000 IPRO and NS 7209 IPRO cultivars were more productive. Despite of the good yielding, the NS 7000 IPRO cultivar contributed in large scale to G x E interaction when considering maturity groups. Due to small coincidence in the ranking of the different strategies, it is possible to infer that the use of only one agronomic trait might not be favorable to classify the superior cultivars.