Tamanho de amostra para caracteres morfológicos e produtivos de Crotalaria juncea

• Autor(a) principal: Schabarum, Denison Esequiel
• Data de Publicação: 2016
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/11641
• topic: Crotalaria juncea L.; Dimensionamento amostral; Planejamento experimental; Sample size; Experimental design; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA

The sunn hemp (Crotalaria juncea L.) is a legume used in crop rotation, for biomass formation, biological nitrogen fixation and control of nematodes. More and more studies are being conducted with this culture and an estimate of adequate sample size becomes very important. Thus, the objective of this study was to determine the sample size to estimate the average morphological and productive traits sunn hemp and verify the variability of the sample size between traits and evaluation periods. And then performing two uniform trials. Both trials were conducted in the field, in the 2014/2015 crop, with sowings in october and december, assessing 300 plants per season. The evaluations of morphological traits (plant height and number of leaves) were held weekly throughout the cycle. The evaluations of the two other morphological traits (stem diameter and root length) and productive traits (fresh and dry matter of leaf, stem, root, the aerial part and total) were evaluated at flowering plants. For each character calculated the sample size for the confidence interval semiamplitudes (estimation errors) equal to 2, 4, ..., 20% of the average estimate. To estimate the average plant height and number of leaves there is variability of the sample size between the stages of development of culture, being necessary sample size at the beginning of the cycle. There is variability between the morphological traits, and the number of leaves require a larger sample size than the other traits. The sample size is larger for the production when compared to morphological traits. There is variability in sample size between the sowing dates. To estimate the average morphological traits with a maximum error of 10% and 95% confidence level should be assessed at least 70 plants and productive traits with a maximum error of 20% and 95% confidence level is due evaluate at least 101 plants.