Google launches WeatherNext 3, adding satellite data for hourly AI weather forecasts
Google has unveiled version 3 of its WeatherNext AI weather model, the latest upgrade in its push to make machine‑learning forecasts as accurate as traditional physics‑based systems while using far less compute. The new release incorporates raw satellite observations, cutting the lag between real‑time conditions and forecast generation and enabling hourly updates instead of the previous six‑hour…
Key points
- WeatherNext 3 adds raw satellite data, enabling hourly AI forecasts
- Spatial resolution and model size increased, with engineering tweaks to limit compute
- Separate ML model now provides satellite‑based precipitation estimates
The upgrade also boosts spatial resolution and expands the model’s size, prompting engineering tweaks to keep computational costs in check. A dedicated satellite‑based precipitation model now runs alongside the primary system, giving users multiple rainfall forecasts. By moving beyond reliance on reanalysis data—global snapshots that blend measurements and are refreshed only every six hours—WeatherNext 3 aims to capture more granular atmospheric details, improving short‑term prediction accuracy.
Google details the changes in a white paper, positioning WeatherNext 3 as a bridge between AI‑driven efficiency and the richness of conventional weather modeling, potentially reshaping how meteorological services deliver timely forecasts.
Update to Google’s AI weather model improves forecast accuracy
Ars Technica AI · 8 September 2026
Google is one of the major players in AI (meaning machine learning) weather forecast model space. The models it and others generate have their strengths and weaknesses, but the main advantage is that they can have forecast performance similar to traditional models while requiring far less computing horsepower to run. That means they can be run more frequently.
Google recently released version 3 of its WeatherNext model, with the biggest change being that it now ingests some satellite weather data, shortening the lag time between current weather conditions and generating a new forecast. The update is detailed in a white paper.
Reanalysis
Many weather models make use of what’s called a “reanalysis,” which is a sort of model of its own. Reanalyses take in all kinds of weather data and combine them into a single, consistent global snapshot of the atmosphere. That requires that they provide estimates for conditions over locations without real-world measurements, because weather forecast models need to work with a global picture.
Nearly all AI weather models have been relying entirely on reanalyses, with machine-learning algorithms training on global reanalyses and spitting out a weather map in the same format. There are some compromises there—the raw data sources themselves may contain some information that gets lost in the reanalysis blender, and these global snapshots are generally only produced every six hours.
Traditional weather forecast models often also take in other raw data, capturing as much information as possible to accurately represent the current state of the atmosphere so the model can use physics to simulate conditions forward. WeatherNext 3 now does some of this as well, adding in weather satellite data and upping the forecast frequency to hourly as a result.
There are some other changes, too. The spatial resolution has been increased, and the machine-learning model is larger, prompting some process tweaks to limit the increased computational demands. They also added a separate machine-learning model trained on satellite-based precipitation estimates, meaning there are multiple precipitation forecasts available.
This text was published by Ars Technica AI and written by Scott K. Johnson. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
2 sources- Weatherwatch: AI model beats standard methods at predicting cyclones Press · The Guardian AI ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us.
Comments
via GitHub Discussions