Magdalena Mair
University of Bayreuth, Junior Research Group Statistical Ecotoxicology, BayCEER
Germany
Of honeybees and water fleas: Data-driven approaches to environmental risk assessment
Protecting the environment and its organisms from harmful pollutants is challenging, as only a few surrogate species can be tested experimentally. New Approach Methodologies (NAMs), including computational tools and predictive modeling, are becoming increasingly important in ecotoxicological research and regulation. Automated experiments reduce manual workload, while machine learning enhances toxicity predictions for untested chemicals across species. Alongside these advances, a conceptual shift in statistical methods now emphasizes demonstrating the absence of adverse effects, raising new concerns about test reliability and statistical power. Using honeybees and water fleas as surrogate species, I will explore how statistical approaches may affect conclusions, highlight the role of open data and structured databases, demonstrate progress in automatized image-based data acquisition and high(er)-throughput testing, discuss the unique challenges of assessing microplastics, and propose how more flexible predictive frameworks can improve environmental hazard assessments.
Jonathan Jeschke
Leibniz Institute of Freshwater Ecology and Inland Fisheries (IGB)
& Freie Universität Berlin, Institute of Biology
Germany
Hi Knowledge: network models for ecological synthesis and knowledge transfer
In the current era, an unprecedented amount of data and information is available to us in principle, yet these data and information are largely disconnected and trapped in silos. This is true among and even within scientific disciplines including ecology, where there is often a disconnect between researchers working empirically in the laboratory or the field versus those working on theory and models. And different ecological research fields focus on different drivers of biodiversity and global change, making integration across fields challenging. In the Hi Knowledge initiative (www.hi-knowledge.org), we have developed and applied approaches to improve synthesis and knowledge transfer. For example, we have created interactive, structured networks of the major hypotheses of invasion biology and urban ecology, and have connected them with empirical studies. I will present these approaches, highlighting their applicability across research fields, and will outline plans for the future including opportunities for collaboration.
