Abstract
Solar radiation is the prime energy source on Earth. It reaches any object in the form of electromagnetic radiation that may be absorbed, transmitted or reflected. The magnitude of these optical processes depends on the optical properties of each object, which in the case of plants relate to their biochemical and structural traits. These plant phenotypic traits result from gene expression underpinned by an individual’s genotype constrained by phylogeny, the environment the individual is exposed to, and the interaction between genotype and the environment. Remote observations of plant phenotypes across space and time may thus hold information about the composition and structure of genetic variation, if a link between spectral and genetic information can be established. This dissertation encompasses studies linking information derived from imaging spectrometer acquisitions under natural conditions with in situ collected information about genetic variation within a tree species, the European beech Fagus sylvatica. It presents the correlation between spectral and genetic information by sequentially expanding temporal, spatial and genetic aspects, and simultaneously accounting for environmental contexts that impact gene expression. By evaluating spectral-genetic similarities across decadal airborne imaging spectrometer acquisitions and accounting for spectral phenotypes and whole-genome sequences of tree individuals from across the species range, the studies provide a proof that observed reflectance spectra hold information about genetic variation within the species. Further, by accounting on uncertainties of spectral measurements and deriving genetic structure of the most abundant tree species in Europe, the dissertation advances the current remote sensing approaches and the knowledge on intraspecific genetic variation. The studies focus particularly on the genetic relatedness between the trees of the test species, whereas the acquired data may allow to establish direct associations between genes and spectral features. The methods used may be expanded to other tree species or applied to spectral data acquired by upcoming spaceborne imaging spectrometers, which overcome current spatiotemporal limitations of data collection, and demonstrate further paths towards the association of genetic variation with variation in spectral phenotypes. The thesis presents the potential of spectral derivation of intraspecific genetic variation within tree species and discusses associated limitations induced by spectral, temporal, spatial and genetic scopes of analysis. This sets a stage towards establishing a means of remote observations of spectral signatures to contribute to monitoring biological variation at the fundamental genetic level, which correlates with ecosystem performance and is an insurance mechanism for populations to adapt to global change.
Czyz, Ewa A