Seleção para vários caracteres agronômicos e reológicos em multiambientes no melhoramento de trigo

Abstract

Wheat cultivation has been growing with the expansion of cultivation in the Brazilian cerrado region, however major challenges are exposed. Mostly, the cultivars currently recommended for this region were developed in breeding programs conducted in southern Brazil. As there are notable macro environmental differences, the development and advancement of breeding populations, as well as the assessment of adaptability and phenotypic stability of cultivars under specific conditions, becomes relevant. This thesis was structured in two articles, which aim to: i) the identification and selection of superior segregating populations in a wheat breeding program through the estimation of mean components; ii) the study of adaptability and phenotypic stability of commercial wheat cultivars recommended for the region of Minas Gerais with a focus on agronomic traits and flour quality. For the first article, wheat populations were generated from a diallel involving 16 lines, evaluated for plant height, grain yield and thousand-grain weight. In 2019, 70 F2 populations and three controls (the cultivars BRS 264, BRS 404 and TBIO Sintonia) were evaluated in an experiment in a partially replicated design (p-rep) at the experimental station of the Agricultural Research Company of Minas Gerais (EPAMIG). In the 2021 harvest, the F3 populations were evaluated in an experiment conducted at the Center for Scientific and Technological Development in Agriculture at UFLA (CDCTA). The experiment was also carried out in a p-rep design. The analyzes were carried out by season and combined with common controls and multivariate analysis using the Z selection index. From the adjusted means, the genetic components of the mean were estimated. The F test was significant for all traits, except for productivity in the F3 generation. The accuracies obtained are considered high, except for productivity in the F3 generation. The selection accuracy when using the Iz index was 43%, considered moderate. Promising populations were identified based on m + a’. For the study of adaptability and stability, 11 commercial wheat cultivars were evaluated in different growing environments in the state of Minas Gerais. Some agronomic characteristics were evaluated: grain yield, weight of a thousand grains and plant height and flour quality characteristics: falling number, hectoliter weight, extraction, elasticity index, P/L ratio, gluten strength and luminosity. Analysis of variance was carried out for all characters and, subsequently, the Scott-Knott test was carried out. Adaptability and stability analysis were carried out using the GGE biplot and Z index with radar graph methods. Differences were identified between cultivars in terms of adaptability and, above all, in terms of stability. It was also possible to distinguish similar environments that can be grouped into mega environments.