Open Data Science Europe Workshop 2021

The presentation will show results of modeling species distribution maps for both potential and actual natural vegetation through spatiotemporal machine learning using a data-driven, robust, objective and fully reproducible workflow. The presentation will focus on the benefits of using ensemble machine learning for species distribution modeling to capture patterns of niche changes in both space and time: yearly (from 2000 to 2020) probability distribution maps for both potential and actual natural vegetation will be presented for forest tree species that live in different climatic conditions across Europe. The high spatial (30 m) and temporal (1 year) resolution of the outputs should allow us to enhance and better understand the patterns of niche change.

Climate change is expected to have strong impacts on forest tree species, since climate influences their physiology, hence phenology and distribution. Species distribution modeling (SDM) is defined as the prediction on geographic distribution of vegetation and its relationships with environmental variables. A spatiotemporal framework is a partition of space and time where observations can be indexed by spatial locations and time stamps.
Potential Natural Vegetation (PNV) is the hypothetical status of vegetation cover in equilibrium with climate, natural disturbances, lithology and hydrologic conditions at a particular location if not impacted by human activities, while Actual Natural Vegetation (ANV) is a narrower environmental space that considers intra and interspecific competition for resources: it can be interpreted as the fundamental niche of a species constrained by biotic interactions, dispersal, but also the realized environment.

The talk will then end with an open discussion on the interpretation of the results, especially on the gaps in range covered by each species between its potential and actual distribution: causes and solutions (i.e. a reforestation program) will be discussed.