OpenPrioBio

OpenPrioBio - Early prioritisation of areas for the biodiversity by fusing multimodal earth observation data with open geodata and AI

As a result of climate change and the encroachment of humans into various ecosystems, biodiversity is increasingly under threat worldwide.
The identification areas in need of protection to promote biodiversity is therefore of central importance in order to protect the diversity and composition of species, ecosystems and their ecosystem services for humans and thus limit the effects of climate change.

As part of the EU Biodiversity Strategy for 2030, a programme was therefore developed under the title „Mehr Raum für die Natur in unserem Leben", the goal was set to protect 30% of aquatic and terrestrial areas in Europe by 2030. To date, however, there is no concept agreed between the stakeholders as to where in Europe or Germany the additional protected areas are to be designated. In addition, due to extreme weather events and natural hazards such as drought, fire, storms, flooding insect calamities or the invasion of invasive species, habitats with high biodiversity are shifting. Until now, protected areas have generally prioritised on the basis of a one-off biological inventory and without taking into climate change and human interventions (Welch et al. 2020). In order to achieve the objectives of the Green Deal, the current protected area landscape must be reviewed and, if necessary be adapted and new areas of high conservation value identified.

The research project "OpenPrioBio - Early prioritisation of areas for the biodiversity by fusing multimodal earth observation data with open geodata and AI" is therefore concerned with the identification of areas in need of protection, taking into account climatic changes and changing anthropogenic influences. The aim of the project is to develop an innovative, transferable method for prioritisation of areas in need of protection to promote biodiversity through the fusion of multimodal earth observation data with open geodata and AI. The objectives are to be achieved by using two methods from the field of AI, reinforcement learning and reinforcement learning and explainable AI (XAI).

More detailed information on the project and the relevant contacts can be found at the following link: https://www.tu.berlin/geoinformation/forschung/projekte/laufende-projekte/openpriobio