IDENTIFICATION OF CORRELATION PATTERNS IN STRUCTURAL ANALYSIS DATA AND IN THE DYNAMICS OF SOYBEAN REFLECTION INDICES

The article presents the results of the study of breeding material on three-year data of structural analysis of soybean plants, interrelations of breeding traits among themselves and their correlation dynamics with different indices of reflection (NDVI, NDRE, ClGreen, GNDVI, MCARI) on the fields of FSC LGC during the growing season. High stability of manifestation of traits associated with the variety was confirmed: weight of 1000 grains and protein and fat content in grains. The traits of height and number of branches on the plant had an average correlation between varieties by years. Other structural indices studied showed a weak level of cohesion among varieties between years. It was noted that the studied varietal samples, having stable by year manifestations of traits, had weak or very weak correlation of these characteristics among themselves. Conversely, traits that were unstable across years often had high correlations with each other. Visual analysis of the dynamics of changes in the correlation relationship of structural analysis indicators and reflection indices showed that during the vegetation of soybean plants there is a smooth change in the values characterizing the relationship. The same dynamics of correlation of traits with high cohesion between each other and by years is observed. For structural indices that stably show correlation relationship by years, two optimal phases fixed by different vegetation indices can be distinguished – V-3/V-4 and R-1. At this time, the studied structural indices have the highest (lowest) correlations with reflection indices. All studied vegetation indices showed a decrease in correlation with yield during development and maturation of soybean plants. At the same time, grain quality (fat and protein) is marked by MCARI index in flowering phase when it has the highest average correlation values.