Análisis estadístico funcional de la reflectancia obtenida mediante imágenes satelitales y espectroradiometro del cultivo de Quinua en Colta

Abstract

The purpose of this work is to identify the behavior of the reflectance of the quinoa crop in Colta, from the values collected through a spectroradiometer that are provided by the Group of Alternative Energies and Environment (GAEE) of the Polytechnic of Chimborazo. In the study, 14 alterations of the seed were considered, among them 8 lines and 6 plots, for each modification, four leaves and one flower of each of the three plants were analyzed, the functional data methodology was used: B-splines smoothing techniques. cubic, detection of atypical functions, descriptive statistics, Principal Components (PC) and ANOVA, the satellite image SENTINEL 2B of the research plot was also identified, to know if the distribution of the curves present any similarity with the spectral signature of the experimental analysis. Due to the fact that the canton is highly susceptible to meteorological variables, it was determined if there is a relationship with the modifications of the crop. A total of 23 functional bases were obtained, the two functional PCs explain more than 80% of the total variability, in addition to the fact that the first two plots resemble quinoa lines 4 and 5 with respect to the flowers, the functional ANOVA presented characteristics similar between sheets with a confidence level of 98%. And the correlation analysis indicated that there is a relationship in at least one meteorological variable. It is concluded that the spectral signature of the quinoa plant is considered healthy and it is better adapted to humid climate conditions with minimum temperatures. It is recommended that in the investigation different techniques should be evaluated with functional statistical analysis methodology.