Upscaling modelisation of climate-induced stone pine colonization into the alpine zone via airborne remote sensing and deep learning

Climate change is significantly altering the functioning of ecosystems and biodiversity patterns and is causing profound modifications of the landscape. Monitoring design and management strategies that account for the scale, pace, and complexity of climate-induced impacts on the environment require robust projections of the potential future distributions of species and ecosystems which is the ultimate aim of this research project.

Vegetation distribution in the Alps is directly related to aspects such as geomorphic processes, water availability, and plant dispersal modes (e.g. animals or wind) as well as indirectly to human activity from agriculture to leisure, and tourism. Nevertheless, recent global warming trends have begun to affect the limits and spatial structure of vegetation colonies in the Alps. While these changes can be monitored locally, a regional-wide characterization is needed to accurately model and forecast their evolution. To address this need, the research team will develop a model of species distribution, linking in situ observations with Earth observation data. In particular, the interdisciplinary team will leverage existing species characteristics and environmental parameter observations and link them spacially to species-specific spectral signatures obtained via a high resolution, airborne imaging spectrometer (AIS). As the foundation of this research project, they will develop an AIS processing framework, including geometric and radiometric calibration methodologies, to generate analysis-ready-data.

The stone pine (Pinus cembra L.) has been selected as the study species for this project as it is one of the most vulnerable treeline species to climate change in the Alps. Moreover, the stone pine is one of the species with greatest potential to transform the alpine landscape in the coming pine is one of the species with greatest potential to transform the alpine landscape in the coming decades.