Training machine learning algorithms: New ways to analyze and predict patterns and correlations in cross-scale biodiversity data (BioTrain)

The BioTrain project will use powerful machine learning algorithms based on biodiversity and environmental data to make predictions about the functionality of ecosystems. Furthermore, options for action to promote functionality and an early warning system to prevent negative environmental effects will be established. To this end, a data scientist affiliated with the Anhalt Center for Data Science will collaborate with experts from two different fields of biodiversity research.

Project priorities

The “Mobile Links” section deals with the movement of organisms, which functionally influences the composition of biotic communities and thus biodiversity. The aim is to develop predictive models for the influence of grazing animals (mobile links) on biotic communities of different taxa and thus the resilience of open landscapes across different spatial and temporal scales.

The second area, “Microbial Communities,” deals with soil and rhizosphere microbiomes in arable soils. The aim is to identify the main factors influencing soil suppressiveness for the integrated control of phytopathogens in soils under different agricultural management practices and to develop strategies for promoting soil health for sustainable soil management.

Project region

Sachsen-Anhalt