Bioelectrical pattern discrimination of Miconia plants by spectral analysis and machine learning

Abstract

We have carried out an in loco investigation into the species Miconia albicans (SW.) Triana and Miconia chamissois Naudin (Melastomataceae), distributed in different phytophysiognomies of three Cerrado fragments in the State of Sao Paulo, Brazil. We\xa0characterized their oscillatory bioelectrical signals and\xa0asked whether these signals\xa0show distinct spectral density. The experiments provided a bank of bioelectrical amplitude samples, which were analyzed in the time and frequency domain. On the basis of the power spectral density (PSD) and machine learning techniques, analyses in the frequency domain suggested that each\xa0of these species has a\xa0unique biological pattern. Comparison between their oscillatory behavior showed bioelectrical features, and both species displayed a bioelectrical pattern, while environmental factors\xa0also influence this pattern. From the point of view of experimental Botany, new questions and concepts could be formulated to advance\xa0the understanding of the interactions between the communicative nature of plants and the environment. The results of this on-site technique represent a new methodology to acquire non-invasive information that might be associated with physiological, chemical, and ecological responses of plants.