Metodologia de germinação, morfologia por análise de imagem e caracterização de espécies por espectroscopia no NIR de sementes de Amaranthus

Abstract

The Amaranthus genus is widely known as an invasive plant competing with agricultural crops, however, it has multiple uses that can contribute to food and nutritional security, medicinal and animal feed. Therefore, it is interesting to encourage knowledge about species and the consumption of these plants that can contribute to the diverse needs in the world. It is worth mentioning that the production of seeds of these species remains one of the major bottlenecks. Researchers' efforts are needed to expand knowledge about these seeds so that they can be made available with adequate quality for commercial production. The use of image analysis techniques (X-ray, GroundEye) and other innovative technologies such as near infrared spectroscopy (NIR) for the characterization of the species can contribute in time and expense in the study of research of these species that are considered extremely potential. Article one the objective of the present study was to determine the methodology for the test germination and morphometry seed of different Amaranthus species. The seed was available in completely randomized a 4 x 2 x 6 factorial arrangement, consisting of the combinations of four species (A. viridis L., A. retroflexus L., A. hybridus L. and Amaranthus sp.), two substrates (paper and vermiculite) and six temperatures (15, 20, 25, 30, 15-25 and 20-30 °C). The germination test can be conducted at a temperature of 30 °C for A. viridis, A. hybridus, Amaranthus sp. and for Amaranthus retroflexus at 25 °C. The paper substrate provides adequate conditions for germination of different species of Amaranthus. The first germination counts at 5 days and the final seed germination test count at 12 days. Morphometry varies according to species. However, the characteristics of sphericality and area provide differentiation between them. Article two the aim of this study was to distinguish between seeds and oils of different species of Amaranthus, using near infrared spectroscopy. Three species were used: A. viridis L., A. hybridus L. and Amaranthus cf. (commercial). The samples were subjected to analysis using near infrared spectroscopy (NIR) and the spectra were generated using the FT-NIR spectrometer. The spectra were subjected to discriminant analysis by partial least squares (PLS-DA) and principal component analysis (PCA). Through PCA for both seed and oil it was possible to differentiate the species of Amaranthus. By PLS-DA it was possible to predict the classes of species with a high degree of correct classification, obtaining 96.67% of correct classifications for seeds and 98.89% for oil.