ResearchGoogle DeepMindConfirmed
Story 14 of 15 in the Friday 13 November editionGoogle DeepMind opens PARDIG and SatWeave datasets from WeatherNext 3
Released on Google Cloud on 13 November 2026 with historical archives and near-real-time streams, PARDIG and SatWeave open the radar and geostationary satellite inputs behind WeatherNext 3.
Edited by Ashutosh Shrivastava
14 Nov 2026, 09:50 IST · 1 min read
Google DeepMind openly released the two foundational observational datasets behind its hourly WeatherNext 3 global forecasting model on Google Cloud on 13 November 2026: PARDIG, an AI-driven global precipitation reanalysis delivering 10 km maps every 15 minutes, and SatWeave, a seamless 17-channel mosaic fusing five geostationary satellites every 60 minutes across historical archives and near-real-time streams.
Study details
- PARDIG specification
- AI-driven global precipitation reanalysis at 10 km spatial resolution updated every 15 minutes
- SatWeave specification
- Global mosaic of 5 geostationary satellites across 17 spectral channels updated every 60 minutes
- Cloud access modes
- Historical reanalysis archives plus near-real-time operational streams hosted on Google Cloud
Release details & context
Direct satellite assimilation in WeatherNext 3
Unlike earlier numerical and AI weather models that wait on multi-hour data-assimilation cycles, WeatherNext 3 ingests raw geostationary imagery directly to produce global hourly ensemble forecasts down to 5 km resolution for station-targeted temperature and humidity and 10 km for surface wind.
Operational WeatherNext 3 forecasts already feed Google Search, Google Maps, Gemini, BigQuery, Earth Engine and Google Cloud Storage alongside renewable-energy radiation and cloud-cover planning.
Sources
Google DeepMind release timeline (9 launches tracked)
- Google DeepMind publishes economic framework for multi-agent AGI regimes
Google DeepMind modelled four AI capability regimes, finding that multi-agent coordination thresholds trigger abrupt jumps in firm concentration and wages.
- Google DeepMind uses AlphaProof Code and Gemini 4 Carbon to formally verify Signal
Google DeepMind’s AlphaProof Code generates 6 million lines of Lean proofs in 4 weeks to verify Signal down to compiled machine code.
- Google DeepMind unveils Gemini Robotics 2.5 for dexterous manipulation and factory kitting
Google DeepMind rolls out Gemini Robotics 2.5, reaching 96% autonomous kitting accuracy on Apollo 2 humanoids in Mercedes-Benz factory pilots.
- Google DeepMind releases AlphaGenome 2 API for single-cell DNA variant prediction
Google DeepMind launches AlphaGenome 2 with custom scRNA-seq conditioning, imputed tracks and CpG methylation scoring across 1 Mb DNA windows.