Environmental data, Earth system variability, and data-constrained future projections – research agenda of the Environmetrics group

Abstract: As environmental change increasingly shapes political, economic, and societal decisions, the question is no longer whether data matter - but how reliably heterogeneous data sources and process understanding can support trustworthy, evidence-based decision making. Environmetrics is the subdiscipline of statistics that develops methods for analyzing environmental and climate data. It comprises a wide range of methods due to the broad range of scientific questions and data types, ranging from satellite observations and instrumental measurements to counts of fossilized pollen or animal abundances. One common feature among most scientific questions is that a lack of real-world controlled experiments requires the integration of multiple data sources and complex process-based numerical models to provide robust results.

In his talk, JProf. Nils Weitzel introduces the research agenda of his newly established Environmetrics group at TU Dortmund University and the Research Center Trustworthy Data Science and Security (RC Trust). After a short overview over the research field, he will present ongoing work to characterize the spatio-temporal structure of climate variability, which emerges from the interaction of external influences and internal dynamics. He combines instrumental observations, natural archives like sediment cores, and climate models to show that state-of-the-art climate models faithfully reproduce global mean temperature variability across timescale but underestimate regional variability on decadal and longer timescales. This result implies a misrepresentation of the spatial structure of climate variability in climate models, with potential implications for projections of future regional climate change.

In the final part of the talk, he will give an overview over other research topics that the Environmetrics group is currently working. The talk will close with a discussion of envisaged steps towards improved future Earth system projections, that utilize new constraints from past environmental changes, and potential collaboration topics.