Reproduce “An AI-Based OSSE Framework to Evaluate Learned Physical Consistency: A Multi-Model Study Using FourCastNet, Aurora, Prithvi WxC, and GraphCast” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yash Diggikar, Milton Halem, Aishwarya Raman
The authors introduce an idealized AI observing-system simulation experiment testing whether correcting one atmospheric field improves others. After a 30-day free run, they replace temperature, wind, or geopotential height with reanalysis fields every six hours for 60 days. At 2°N and 500 hPa, temperature insertion reduces zonal-wind RMSE by about 52% for Aurora; results vary by model and do not establish real-world observing-system impact.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.