Modelos de crescimento para cultivares de capim-sudão em épocas de semeadura

• Autor(a) principal: Pezzini, Rafael Vieira
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/20747
• topic: Sorghum sudanense (Piper) Stapf; Cobertura de solo; Modelos não lineares; Soil cover; Nonlinear models; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA

The objectives of this work were to fit the Gompertz and Logistic models to describe the growth of morphological and productive traits as a function of accumulated thermal sum and accumulated solar radiation, comparing the adjustments and indicating the best to describe the growth of two sudangrass cultivars in four sowing seasons. Eight uniformity trials were conducted with Sudan grass, composed by the combination of two cultivars (BRS Estribo and CG Farrapo) sown in four seasons (20/12/2016, 20/01/2017, 02/02/2017 and 24/02/2017). The sowing was done in rows, spaced 0,4 m between rows. Each trial occupied an area of 9 m × 16 m (144 m²). The evaluations were performed three times a week from 15 days after the emergence of the plants. Each evaluation consisted of the collection of five plants of each trial, for measurement of the plant height and stem length and the weighing of the fresh matter of aerial part and dry matter of aerial part. The Gompertz and Logistic models were fitted for each trait as a function of accumulated thermal sum and accumulated solar radiation. The parameters were estimated, determined the confidence interval of the parameters, the critical points of the growth curves and the adjustment quality indicators. The intrinsic nonlinearity and the nonlinearity of parameter effect were also quantified. The independent variables accumulated thermal sum and accumulated solar radiation can be used to fit the models. Both models are suitable for describing the morphological and productive traits of the of BRS Estribo and CG Farrapo cultivars. However, the Logistic model is more indicated because it presents greater accuracy.