Google DeepMind opens SatWeave and PARDIG datasets behind WeatherNext 3
ymatias · x · 2026-11-16
Google DeepMind released two foundational datasets behind WeatherNext 3: SatWeave, a global mosaic blending 5 geostationary satellites into 17 spectral channels with a 10+ year archive and live feed; and PARDIG, a 15-minute 10 km global precipitation dataset (1999–2026 archive, 5-hour latency) improving light-to-moderate rain detection by >30% over NASA's IMERG-Final. Both are available via Google Cloud Storage and Google Earth Engine, with an accompanying research paper.
Related event: Google Open-Sources SatWeave and PARDIG Datasets Behind WeatherNext 3(2 posts)→
More from Research
- OSWorld-Persona: a 3,259-label benchmark for evaluating when agents should proactively help — EchoShao8899 · 2026-11-16
- Fable 5.2 reproduces outdated MazeBench solutions, sparking contamination suspicion — patience_cave · 2026-11-16
- Study asks whether people want AI advisors aligned on decisions, reasons, or both — MaartenSap · 2026-11-16
- Schmidhuber Publishes New Survey: Machine Learning Is the Science of Credit Assignment — SchmidhuberAI · 2026-11-16
- Andrew Wilson: AI research beyond black-box prediction could yield the next Transformer — andrewgwils · 2026-11-16
- Perplexity Launches Frontier Research to Openly Advance Models, Data and Algorithms — andrewgwils · 2026-11-16