Training Machine Learning Algorithms: New Pathways to Analyze and Predict Patterns and Relationships in Cross-Scale Biodiversity Data (BioTrain)

The BioTrain project utilizes powerful machine learning algorithms based on biodiversity and environmental data to forecast ecosystem functionality. Additionally, it will establish management options to promote ecosystem functionality and an early warning system to prevent negative environmental effects. To achieve this, data scientist based collaborate with specialists from two different areas of biodiversity research.

 

The work package "Mobile Links" focuses on the movement of organisms, which functionally influences community composition of different organism groups and, thus, biodiversity. The goal is to develop predictive models for the influence of grazing animals (mobile links) on communities of various taxa and, consequently, the resilience of open landscapes across different spatial and temporal scales.

The second work package, "Microbial Communities," deals with soil and rhizosphere microbiomes in arable soils. The objective is to identify the key factors affecting soil suppressiveness for integrated control of phytopathogens in soils under different agricultural management practices and to develop strategies for promoting soil health for sustainable soil management. 

Project Lead: Prof. Dr. Christina FischerProf. Dr. Korinna Bade, Prof. Dr. Wilfried Rozhon

 

Project Management: Roi Hendler

Project Duration: 12/2022 - 11/2025

Funding Code: 16DKWN019