Integração de variáveis climáticas via decomposição em valores singulares para predição da interação genótipo por ambiente em trigo
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidade Federal de Viçosa
Abstract
O trigo (Triticum aestivum L.) é uma das culturas agrícolas mais importantes para a segurança alimentar global e para a economia de diversos países. O desempenho dos genótipos é influenciado por fatores genéticos e ambientais, bem como pela interação entre esses componentes. Assim, compreender a interação genótipo × ambiente (G × E) é essencial para o desenvolvimento e a recomendação de cultivares mais produtivas, adaptadas e resilientes. Entretanto, a elucidação dos mecanismos que estruturam essa interação requer a incorporação de covariáveis climáticas capazes de caracterizar quantitativamente os ambientes de cultivo e explicar os fatores ambientais associados ao desempenho diferencial dos genótipos, especialmente em um cenário de crescente variabilidade climática. Embora métodos tradicionais permitam detectar e quantificar a interação G × E, os ambientes são frequentemente tratados como fatores categóricos, limitando a interpretação biológica dos mecanismos ambientais associados ao desempenho dos genótipos. Nesse contexto, este estudo propôs uma abordagem gráfica baseada na decomposição em valores singulares (SVD) de covariáveis climáticas para caracterizar quantitativamente os ambientes e interpretar os padrões da interação G × E em trigo. Foram utilizados dados de 42 genótipos de trigo avaliados em dez ambientes, definidos pela combinação de dois anos agrícolas e cinco locais do estado do Rio Grande do Sul, em delineamento de blocos casualizados com quatro repetições. As covariáveis climáticas, incluindo temperatura do ar, umidade relativa e precipitação, foram obtidas a partir do banco de dados NASA POWER, considerando os anos agrícolas e as coordenadas geográficas de cada ambiente experimental. A interação G × E foi investigada por meio do modelo AMMI, enquanto a adaptabilidade e a estabilidade dos genótipos foram avaliadas utilizando o índice WAASY. Adicionalmente, a abordagem baseada em SVD permitiu caracterizar os ambientes e identificar gradientes climáticos associados aos padrões de produtividade. Os resultados evidenciaram que os ambientes E6 e E8 apresentaram maior produtividade e foram caracterizados por condições associadas à maior umidade relativa do ar e à precipitação, enquanto ambientes menos produtivos estiveram associados a variáveis relacionadas à temperatura. A abordagem proposta permitiu identificar gradientes climáticos, agrupamentos de ambientes semelhantes e fatores meteorológicos relacionados ao desempenho diferencial dos genótipos. Conclui-se que a integração de covariáveis climáticas via SVD amplia a interpretação biológica da interação G × E ao permitir a caracterização quantitativa dos ambientes experimentais, evidenciando gradientes climáticos e associando padrões de produtividade a variáveis específicas de temperatura, umidade relativa e precipitação, fornecendo subsídios para estratégias de seleção e recomendação de cultivares mais adaptadas às condições ambientais. Palavras-chave: Palavras-chave: adaptabilidade e estabilidade fenotípica; índice WAASY; caracterização ambienta; SVD.
Wheat (Triticum aestivum L.) is one of the most important agricultural crops for global food security and the economies of several countries. The performance of genotypes is influenced by genetic and environmental factors, as well as by their interaction. Therefore, understanding the genotype × environment (G × E) interaction is essential for the development and recommendation of more productive, adapted, and resilient cultivars. However, elucidating the mechanisms underlying this interaction requires the incorporation of weather covariates capable of quantitatively characterizing growing environments and explaining environmental factors associated with differential genotype performance, particularly in the context of increasing weather variability. Although traditional methods allow the detection and quantification of G × E interaction, environments are often treated as categorical factors, limiting the biological interpretation of the underlying environmental mechanisms. In this context, this study proposes a graphical approach based on singular value decomposition (SVD) of climatic covariates to quantitatively characterize environments and interpret G × E interaction patterns in wheat. Data from 42 wheat genotypes evaluated in ten environments, defined by the combination of two growing seasons and five locations in the state of Rio Grande do Sul, Brazil, were analyzed in a randomized complete block design with four replications. Weather covariates, including air temperature, relative humidity, and precipitation, were obtained from the NASA POWER database, considering the growing seasons and geographic coordinates of each experimental site. The G × E interaction was investigated using the AMMI model, while genotype adaptability and stability were assessed using the WAASY index. Additionally, the SVD-based approach enabled the characterization of environments and the identification of climatic gradients associated with yield patterns. Results showed that environments E6 and E8 exhibited higher grain yield and were characterized by climatic conditions associated with higher relative humidity and precipitation., whereas lower-yielding environments were associated with temperature-related variables. The proposed approach allowed the identification of climatic gradients, clustering of similar environments, and meteorological drivers of differential genotype performance. It is concluded that the integration of climatic covariates via SVD enhances the biological interpretation of the G × E interaction by enabling the quantitative characterization of experimental environments, revealing climatic gradients and associating yield patterns with specific variables such as temperature, relative humidity, and precipitation, thereby supporting breeding strategies for the selection and recommendation of cultivars better adapted to environmental conditions. Keywords: phenotypic adaptability and stability; WAASY index; environmental characterization; SVD
Wheat (Triticum aestivum L.) is one of the most important agricultural crops for global food security and the economies of several countries. The performance of genotypes is influenced by genetic and environmental factors, as well as by their interaction. Therefore, understanding the genotype × environment (G × E) interaction is essential for the development and recommendation of more productive, adapted, and resilient cultivars. However, elucidating the mechanisms underlying this interaction requires the incorporation of weather covariates capable of quantitatively characterizing growing environments and explaining environmental factors associated with differential genotype performance, particularly in the context of increasing weather variability. Although traditional methods allow the detection and quantification of G × E interaction, environments are often treated as categorical factors, limiting the biological interpretation of the underlying environmental mechanisms. In this context, this study proposes a graphical approach based on singular value decomposition (SVD) of climatic covariates to quantitatively characterize environments and interpret G × E interaction patterns in wheat. Data from 42 wheat genotypes evaluated in ten environments, defined by the combination of two growing seasons and five locations in the state of Rio Grande do Sul, Brazil, were analyzed in a randomized complete block design with four replications. Weather covariates, including air temperature, relative humidity, and precipitation, were obtained from the NASA POWER database, considering the growing seasons and geographic coordinates of each experimental site. The G × E interaction was investigated using the AMMI model, while genotype adaptability and stability were assessed using the WAASY index. Additionally, the SVD-based approach enabled the characterization of environments and the identification of climatic gradients associated with yield patterns. Results showed that environments E6 and E8 exhibited higher grain yield and were characterized by climatic conditions associated with higher relative humidity and precipitation., whereas lower-yielding environments were associated with temperature-related variables. The proposed approach allowed the identification of climatic gradients, clustering of similar environments, and meteorological drivers of differential genotype performance. It is concluded that the integration of climatic covariates via SVD enhances the biological interpretation of the G × E interaction by enabling the quantitative characterization of experimental environments, revealing climatic gradients and associating yield patterns with specific variables such as temperature, relative humidity, and precipitation, thereby supporting breeding strategies for the selection and recommendation of cultivars better adapted to environmental conditions. Keywords: phenotypic adaptability and stability; WAASY index; environmental characterization; SVD
Description
Citation
FERNANDES, Carla Galvão. Integração de variáveis climáticas via decomposição em valores singulares para predição da interação genótipo por ambiente em trigo. 2026. 39 f. Tese (Doutorado em Genética e Melhoramento) - Universidade Federal de Viçosa, Viçosa. 2026.
