Análisis de la reflectancia del cultivo de quinua en función de la edad de la planta, mediante estadística funcional

Abstract

This research aimed to identify the reflectance behavior of the quinoa plant (Chenopodium quinoa) taken with a spectroradiometer and associate it with its age. A simple random sampling was done by the Alternative Energies and Environment Group (GEAA) in conjunction with the European Committee for Training and Agriculture (CEFA) and the National University of Chimborazo (UNACH). For the analysis, functional data statistics were used, since it is an excellent tool for the study of continuous variables, a smoothing of B-splines was performed to convert the specific observations into curves, in order to reduce noise and bias, through cross-validation the number of optimal bases was identified. To recognize the way in which the data behave, a descriptive and exploratory analysis was carried out, atypical curves were distinguished that were separated from the analysis; Through a principal components analysis, the reflectance in the plane was represented, with a percentage of explained variability greater than 90% for the 33 samples, patterns were visualized in each of these, which were supported by a functional variance contrast (FANOVA). both in leaves and flowers. It was concluded that there are significant differences between each of the ages of the plant, in the same way in the flowers due to the variety of colors they present caused by their genetic combinations which fluctuate even more due to their physiological maturity; on the other hand, the leaves presented a similar behavior, that is, they are statistically the same. It is suggested to monitor a sample from its first phenological stage to the time of harvest and compare the results with this research.