Erich M. Fischer | Searcharxiv

arXiv2026

AI-Assisted Scientific Assessment: A Case Study on Climate Change

The emerging paradigm of AI co-scientists focuses on tasks characterized by repeatable verification, where agents explore search spaces in 'guess and check' loops. This paradigm does not extend to problems where repeated evaluation is impossible and ground truth is established by the consensus synthesis of theory and existing evidence. We evaluate a Gemini-based AI environment designed to support collaborative scientific assessment, integrated into a standard scientific workflow. In collaboration with a diverse group of 13 scientists working in the field of climate science, we tested the system on a complex topic: the stability of the Atlantic Meridional Overturning Circulation (AMOC). Our results show that AI can accelerate the scientific workflow. The group produced a comprehensive synthesis of 79 papers through 104 revision cycles in just over 46 person-hours. AI contribution was significant: most AI-generated content was retained in the report. AI also helped maintain logical consistency and presentation quality. However, expert additions were crucial to ensure its acceptability: less than half of the report was produced by AI. Furthermore, substantial oversight was required to expand and elevate the content to rigorous scientific standards.

cs.CL↗

arXiv2026

Omega-blocks with spatially compounding extremes over Europe are highly sensitive to remote atmospheric drivers

Omega-blocks can trigger spatially compounding heat-precipitation extremes with severe societal impacts, as seen in September 2023 when a heatwave over France coincided with devastating floods in the Iberian Peninsula and Greece. Although blocking in general has been linked to moist processes in upstream warm conveyor belts (WCBs), it has remained unexplored whether and how upstream WCB activity influences the evolution of omega-blocks and downstream flood-heat-flood impacts. Here, we show that already five days ahead, small differences in the upstream evolution - particularly in WCB outflow regions - distinguish cases that later produce extreme compound events over Europe from weaker ones, even though their large-scale anomalies initially appear similar. We illustrate the distinct evolution in remote locations by analyzing storylines simulated in a fully coupled climate model. Using ensemble boosting, we generate hundreds of physically plausible simulations of omega-prone situations. Lagrangian air parcel tracking reveals that variations in WCB outflow areas can explain differences in upstream precursors and downstream effects over Europe. Our results highlight ensemble boosting as a powerful approach to systematically track dynamical differences along model-based event storylines, important for understanding and anticipating compound extremes striking multiple regions simultaneously.

physics.ao-ph↗