dataset · Global

era5-climate-learn

Hugging Face dataset thainamhoang/era5-climate-learn

Claim this listing

Represent era5-climate-learn? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Data offered docs-verified: ERA5 reanalysis preprocessed for the ClimateLearn benchmark (atmospheric and surface variables, hourly)
Coverage llm-inferred: Global (0.25° grid, per original ERA5 specs in docs) · 1979 to present (per original ERA5 specs in docs; preprocessed subset span not stated)
Resolution llm-inferred: Hourly temporal; 0.25° spatial (~31 km), per original ERA5 specs in docs; preprocessed resolution not stated
Formats llm-inferred: NetCDF
Access docs-verified: Open access under CC BY 4.0 (attribution required); no authentication listed (auth: null)

Use cases llm-inferred: Benchmarking ML climate downscaling models on ERA5 inputs · Testing data-driven weather prediction baselines against ERA5 reanalysis · Historical atmospheric context for ML-based downscaling of temperature or rainfall in climate-change assessments
Audience llm-inferred: hydrologists · climate scientists · data scientists and ML researchers · GIS analysts

Constraints llm-inferred: Hugging Face dataset viewer unavailable; no supported data files auto-detected · Global 0.25° grid (~31 km) is not UK-specific and may be too coarse for local drainage or FRA work · Variable subset, file layout and time span of the preprocessed data are not documented; the stated specs are for original ERA5 · Total size about 179 GB · Citation of both ERA5 and the ClimateLearn benchmark is requested

Verified 2026-10-15 via HF API: exists; 322 downloads/30d, 0 likes.

Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face datasets)