PlantDiv-Spec - Towards robust detection of plant diversity &
management in grasslands’ spectral signals

Runtime: 10/2022 - 09/2024
Team: Dr. Daniel Doktor (PI), Dr. Maksim Iakunin

In the project, spectral and botanical data from grasslands with varying management practices (and corresponding species diversity) were collected for the first time over an entire vegetation period. This enabled a robust parameterisation of a radiative transfer model to theoretically examine whether management-induced changes in plant traits (properties) systematically translate into a remote sensing signal. The results showed that due to simplified model assumptions and standard sampling procedures, a robust derivation of management practices from remote sensing signals is not yet feasible.

The project outcomes provide a basis for further development of radiative transfer models. Specifically, this concerns the inadequate representation of dead (dry) biomass in the simulation models. Additionally, the project results will inform future planning and protocols for plant sampling. There is a need for sampling that accounts for the ratio of green (vital) to dry (dead) biomass, rather than focusing solely on the vital parts of the vegetation. By considering both components, management-induced changes in grasslands can be better identified in the future.