Remote sensing in mapping biodiversity – A case study of epiphytic lichen communities

Palmroos I., Norros V., Keski-Saari S., Mäyrä J., Tanhuanpää T., Kivinen S., Pykälä J., Kullberg P., Kumpula T. & Vihervaara P.
Forest Ecology and Management
538: 120993 [10 p.]
In boreal forests, European aspen (Populus tremula L.) is a keystone species that hosts a variety of accompanying species including epiphytic lichens. Forest management actions have led to a decrease in aspen abundance and subsequent loss of suitable habitats of epiphytic lichens. In this study, we evaluate the environmental responses of epiphytic lichen species richness and community composition on aspen, focusing on the potential of remote sensing by combined hyperspectral imaging and airborne laser scanning to identify suitable habitats for epiphytic lichens. We measured different substrate and habitat parameters in the field (e.g., aspen diameter and bark pH) and by remote sensing (e.g., mean canopy height and tree species composition of the surrounding forest) in the study area in Southern Finland that includes protected and non-protected forest. We used linear regression and the Hierarchical Model of Species Communities (HMSC) to compare how the different parameters explain and predict lichen species richness and community composition, respectively. We show that coarse predictions of epiphytic lichen community composition can be made using parameters extracted from remote sensing data. Estimated mean canopy height, tree density, dominant tree species and tree species diversity of the stand predicted the species community on aspens slightly better than field parameters. Remote sensing variables calculated over a larger area (30 m radius) always outperformed the same variables calculated over a smaller area (10 m radius) in predicting community composition, highlighting the cost-efficiency of remote sensing compared to covering a similar area with on-ground measurements. These results are encouraging for the prospects of using remote sensing data to direct field inventories and to map potential high-biodiversity habitats. Aspen bark pH was the only parameter affecting species richness regardless of whether the forest was protected or not, whereas, interestingly, the effects of tree diameter, height and furrow depth were only significant in protected areas. Our results also underline the importance of protected areas, since they hosted a higher tree-specific number of epiphytic lichen species, and red listed species, than non-protected areas. Keywords: Epiphytic lichen ; Hierarchical model of species communities ; Remote sensing ; Light detection and ranging ; Hyperspectral imaging.
Tuesday, 23 May 2023 00:14