WeatherNext 3

(deepmind.google)

101 points | by matthieu_bl 4 days ago

6 comments

  • hankbond 7 minutes ago
    The exploration page is really easy to use. Having more reliable weather predictions is a genuinely useful product.
  • mrd3v0 16 minutes ago
    "Weather Lab is not available in your country or region at this time." cool...
  • RockstarSprain 43 minutes ago
    Is there a convenient way to use this on iOS? AFAIK Google Search and Maps only show rather basic information.
  • bahmboo 3 days ago
    They even have an interactive world map with all the layers. It's so cool when an announcement comes with the actual goods.
    • ssl-3 26 minutes ago
      I'm not seeing that kind of demo. I just see a video, a bunch of future-tense words about what could be done by someone, and the requisite hype.

      Link?

      • zamadatix 17 minutes ago
        Clicking "Try WeatherNext 3" at the top of the page takes you to the section with different ways to use it.
      • trainingonme 22 minutes ago
        Click on "Explore Weather Lab" at the bottom of the page
  • NostraDavid 1 day ago
    In the energy world, this should be such a boon over the classic NWP (Numerical Weather Prediction; complex ML models), but I've not seen it implementated. Anyone with experience of these models over classic NWP?
    • counters 12 minutes ago
      Why should they be such a boon? High-resolution, regional NWP with fast refresh has been a standard tool for over a decade.
    • BlackRabbit1 27 minutes ago
      A few weather forecast sites in Europe have them on their websites. It's just another forecast you can select and compare in an ensemble.

      Problem is that the WN3 grid is still quite rough (5km) - but a that's a brutal improvement for many places compared to many other global models.

      Quite a few country-scale models go down to a 1-2km grid nowadays. This is very helpful in complex geography like mountains and alleys.

      • counters 5 minutes ago
        > Problem is that the WN3 grid is still quite rough (5km) - but a that's a brutal improvement for many places compared to many other global models.

        That's a pretty apples-and-oranges comparison. One would almost always use a high-resolution regional model if you needed certain details for different forecasting applications like renewable energy.

        It's also worth noting that the 5km outputs are from a model decoder head that was trained against temperature and dewpoint at surface stations. According to the Rasp et al (2026) preprint, this head was designed for continuous sampling; the choice of a 5km grid is arbitrary. What we don't actually know is how well the model handles shocks like a frontal passage or impacts from things like outflow from storms - or even evaporative cooling from precipitation. We are limited to the output that DeepMind publishes; we can't run the model and stress test these things on our own.

        That's all a long way to say that the 5km resolution is (a) limited to temperature fields, and (b) we don't know if the "additional" resolution has any impact whatsoever on the phenomena that one would typically use a mesoscale-resolving forecast for.

      • dist-epoch 5 minutes ago
        You could use this model as another input into the country-scale models.
  • tcumulus 3 days ago
    Paper: https://storage.googleapis.com/deepmind-media/papers/weather...

    TLDR: the input data for the model now also includes real-time observations (satellite and weather stations) on top of the typical (re)analysis data, improving model resolution, run frequency and timestep frequency.

    • dang 53 minutes ago
      We'll put that link in the toptext as well. Thanks!