PulseAugur
EN
LIVE 17:16:44

Google DeepMind open-sources advanced WeatherNext cyclone forecasting AI

By PulseAugur Editorial · [5 sources] ·

Google DeepMind has open-sourced its WeatherNext AI model, which significantly advances cyclone forecasting capabilities. Published in Nature, WeatherNext provides an average of 24 extra hours of lead time for storm track and intensity predictions, matching three-day forecasts with previous two-day accuracy. The model was trained on extensive atmospheric data and historical cyclone records, and can generate 15-day probabilistic forecasts in under a minute. This release includes both the code and model weights, making it available for academic and operational use. AI

IMPACT Accelerates AI adoption in weather forecasting and disaster preparedness.

RANK_REASON Google DeepMind released an AI model with a system card and code/weights.

Read on X — Google DeepMind →

AI-generated summary · Google Gemini · from 5 sources. How we write summaries →

Google DeepMind open-sources advanced WeatherNext cyclone forecasting AI

COVERAGE [5]

  1. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind ·

    We’re open sourcing the code and model weights on @Github, making them freely available for anyone to build on.

    We’re open sourcing the code and model weights on @Github, making them freely available for anyone to build on. This could be for academic purposes, operational forecasting, or developing more specialized, localized models. Explore the research → https://t.co/hyllj4fVKW

  2. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind ·

    During Hurricane Melissa, WeatherNext gave forecasters early predictions of its Category 5 landfall 5 days in advance with 80% confidence.

    During Hurricane Melissa, WeatherNext gave forecasters early predictions of its Category 5 landfall 5 days in advance with 80% confidence. This year, we’re providing 1,000 probabilistic predictions per storm to support forecasters, now accessible via WeatherLab → https://t.co/je…

  3. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind ·

    The model learned from years of everyday global atmospheric data alongside a curated database of almost 5,000 historical cyclones.

    The model learned from years of everyday global atmospheric data alongside a curated database of almost 5,000 historical cyclones. It generates each 15-day probabilistic forecast scenario in under a minute on a TPU. https://t.co/vptoun7H9G

  4. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind ·

    WeatherNext delivers a decade's worth of forecasting progress in a single leap. 📈

    WeatherNext delivers a decade's worth of forecasting progress in a single leap. 📈 On average, 3-day predictions now match the quality that prior models could only provide 2 days out. https://t.co/oHirwymtF7

  5. X — Google DeepMind TIER_1 English(EN) · GoogleDeepMind ·

    Predicting cyclones accurately can help save lives - and every hour of lead time counts.

    Predicting cyclones accurately can help save lives - and every hour of lead time counts. Published in @Nature, our AI model WeatherNext achieves state-of-the-art accuracy in forecasting a storm’s track and intensity, giving us a critical extra 24 hours to prepare on average. 🧵 h…

RELATED ENTITIES

RELATED TOPICS