Is evolution vital for community rescue?

• Applicant: Dr. Ruben Hermann, Ph.D.
• Subject Area: Ecology and Biodiversity of Animals and Ecosystems, Organismic Interactions ; Bioinformatics and Theoretical Biology ; Ecology and Biodiversity of Plants and Ecosystems
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 560549853

Many communities are threatened with their potential collapse due to anthropogenic stressors. Yet, some communities are able to avert their biomass collapse in a detrimental environment, which has been defined as community rescue (CR). This process has been observed in experiments, but up to this point it has not been possible to identify the mechanisms driving CR, which is critical to understand which natural systems are vulnerable to extinction or design effective interventions. While some mechanisms can be identified in small scale experiments, they lack the connection to natural communities. The goal of my project is to infer the mechanisms of CR experiments with semi-natural communities, enabling the transferability to natural systems. I plan four main steps to accomplish this. First, I will develop a perspective article to review the current knowledge on CR and identify various dimensions of biodiversity to consider in CR studies that will help understand it, in cooperation with CR expert Vincent Fugère. This will create a foundation for improved comparisons between CR experiments and observations in nature. Second, I will adapt an existing model to include mechanisms which can reproduce the community dynamics of empirical data, in corporation with the author of the model Ellen van Velzen. The expanded model will increase species and trait diversity to better reflect natural communities, and will be iteratively compared to existing datasets from CR experiments to ensure the model is grounded in a natural system. Third, I will apply this model to the empirical CR data, using Approximate Bayesian Computation (ABC) to test the likelihood that different potential mechanisms drove CR, in cooperation with the statistical modeling expert Jelena Pantel. Different evolutionary and ecological mechanistic hypotheses will be tested with the statistical model to infer the mechanism most likely to have generated the empirical data. The empirical CR data comes from mesocosm experiments in freshwater plankton communities, where herbicide was applied as stressor. Fourth, I will apply this combined theoretical and statistical modelling method to two other CR experiments that applied acidity as a stressor. I can thus determine if a different stressor results in a distinct mechanistic response for CR. ABC allows to infer the state of the community prior the perturbation, which is a unique opportunity to use data-informed model simulations to predict the likelihood of CR and its underlying mechanism, and to test the dependence of CR on different starting conditions of the community. My proposed project will push forward the study of CR from observational to predictive, from post-hoc confirmation to mechanistic understanding, and will also help establish my status as a researcher with leading expertise in incorporating statistical modelling with empirical data to infer previously unknown mechanistic drivers of biodiversity response to anthropogenic change.

• DFG Programme: WBP Fellowship
• International Connection: France
• Host: Professorin Jelena Pantel, Ph.D.