ScienceTrending
Google DeepMind Opens AI Rainfall Datasets With Five-Hour Delay
Google DeepMind has released PARDIG, an AI-powered global rainfall dataset with 15-minute estimates and records back to 1999, under a Creative Commons license.

In brief
- Google DeepMind released PARDIG, an AI-powered global rainfall dataset, on Nov. 13, 2026.
- PARDIG estimates rainfall every 15 minutes at roughly 10-kilometer resolution worldwide.
- Historical records date back to 1999, and both datasets are under a Creative Commons license allowing commercial reuse.
- DeepMind claims more than 30% higher detection skill than NASA's IMERG-Final for light-to-moderate rain.
Google DeepMind Releases Open Rainfall Data
Google DeepMind has opened access to PARDIG, an AI-powered global precipitation monitoring system, researchers and developers can now access historical records and continuously updated data. The company announced the release on Nov. 13, 2026.
The technology combines satellite observations and atmospheric data to estimate rainfall every 15 minutes across a global grid with roughly 10-kilometer resolution. It learns from precipitation measurements collected by NASA's Global Precipitation Measurement satellite and uses additional satellite imagery and atmospheric information to estimate rainfall across the planet.
Unlike conventional monitoring systems that depend heavily on ground-based weather stations, PARDIG can generate estimates over oceans and regions with sparse observation networks.
Dataset Details and Performance Claims
Researchers can access historical rainfall records dating back to 1999, alongside a separate dataset covering 2016 to the present. Google has released both under a Creative Commons license that permits commercial reuse with attribution.
DeepMind said PARDIG achieved more than 30% higher hourly precipitation detection skill than NASA's IMERG-Final dataset when measuring light-to-moderate rainfall. The company evaluated the system against radar networks in the United States, Europe and Japan, alongside observations from more than 36,000 weather stations worldwide.
The dataset also supports Google's WeatherNext 3 forecasting model, which uses PARDIG-generated rainfall information during training.
Limitations and Delivery Delay
Although PARDIG estimates rainfall at 15-minute intervals, Google publishes its near-real-time data in six-hour batches, creating an average delay of about five hours between observation and availability. Cloud Storage delivers them in batches with roughly two to eight hours of delay.
This means emergency services cannot use the dataset as an immediate substitute for live weather warnings. Google also acknowledged that the system can underestimate snowfall in mountainous regions and is less reliable in certain polar areas. The company still classifies PARDIG as experimental and advises users to rely on official meteorological agencies when making public-safety decisions.
Why it matters
- PARDIG provides open, same-day, 15-minute global rainfall fields that researchers can use to study floods and droughts in poorly monitored regions.
- The dataset's claimed skill advantage over NASA's IMERG-Final comes from the dataset's own creators in an arXiv preprint, and independent verification is still pending.
- The roughly five-hour average delivery delay limits its use for rapidly developing emergencies, and Google says it does not replace official warnings.
Sources
- cryptoslate.com: Google opens AI rainfall tracking system that could improve global flood and drought monitoring
- techtimes.com: DeepMind Opens Free AI Rain Data: Five-Hour Latency, Claimed Win Over NASA's IMERG-Final
This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.
Questions readers ask
- What is PARDIG and when was it released?
- PARDIG is an AI-powered global precipitation monitoring system released by Google DeepMind on Nov. 13, 2026.
- What data does the PARDIG dataset include?
- The dataset includes historical rainfall records dating back to 1999 and a separate dataset covering 2016 to the present.
- How accurate is PARDIG compared to other systems?
- DeepMind claims PARDIG achieved more than 30% higher hourly precipitation detection skill than NASA's IMERG-Final dataset for light-to-moderate rainfall.
- Why is there a delay in accessing the data?
- Google publishes near-real-time data in six-hour batches via Cloud Storage, creating an average delay of about five hours between observation and availability.
- Can emergency services use PARDIG for immediate warnings?
- No, the five-hour average delivery delay means emergency services cannot use the dataset as an immediate substitute for live weather warnings.
The day's checked news, by email
One short email a day with the top stories. Free, and you can unsubscribe in one click.
Or follow for the latest news:
How do you feel about this story?
Comments 0
No comments yet. Start the discussion.
More Science stories
Harvard study: US data center emissions match Denmark's, UN launches clean AI pledge
A Harvard study finds the 400 largest US data centers emitted as much heat-trapping gas as Denmark in one year, while COP31 negotiators launched a voluntary clean energy pledge for data centers.
COP31 Draft Calls for Immediate Shift Away From Fossil Fuels
A draft COP31 decision calls for an immediate transition away from fossil fuels and at least $120 billion in adaptation finance for developing countries.

