Quick start: Accessing WeatherNext forecasts

WeatherNext 3 forecast predictions are available across multiple Google Cloud surfaces, including Google Cloud Storage (Zarr v3), Google Earth Engine, and Google BigQuery.

Follow these steps to request access, choose the best surface for your workflow, and run your first query.

Step 1: Request data access

WeatherNext real-time operational datasets are accessible upon request. Submit the access form with your Google Cloud account email to be added to the allowlist:

WeatherNext Data Request Form

Requests are reviewed on a rolling basis and are typically approved within 5–7 business days.

Open Data Request Form →

Account requirements & FAQs

  • What email address should I provide on the form?
    Provide the email address associated with your Google Account (such as your company Google Workspace email or personal @gmail.com) that you use to log into Google Cloud Console, BigQuery, or Earth Engine.
  • Do I need an existing paid Google Cloud contract?
    No. Anyone with a Google Account, including researchers, students, and non-Google Cloud customers can access the data. If you don't have a Google Cloud account yet, you can create a free Google Cloud account in a few minutes.
  • Do I need to submit separate requests for each platform?
    No. Once your Google Account is added to the allow list, you receive access to WeatherNext across Google Cloud Storage, BigQuery, and Earth Engine simultaneously.
  • What terms and licensing apply to the data?
    Real-time data (any data that relates to a time less than 1 hour ago and the future) is governed by the GDM Real-Time Weather Forecasting Experimental Data Terms of Use. Historical data (any data that relates to a time 1 hour ago or more) is licensed under CC BY 4.0. For more details, see Terms of Service and Disclaimers.

Step 2: Choose your surface

Select the platform that matches your tooling, language preference, and data resolution needs:

Surface Best For Data Scope Tooling Documentation
Google Cloud Storage (Zarr) Direct programmatic file access, high performance computing, custom ML models, and raw data extraction. Full fidelity & summary statistics: Raw 64-member ensemble with 3D atmospheric pressure levels in gs://weathernext3_spatial/, and precomputed surface summary statistics in gs://weathernext3_statistics_spatial/. Python (xarray, zarr, obstore) GCS (Zarr) Guide
Earth Engine Planetary-scale geospatial analysis, raster overlays, satellite data fusion, and interactive web mapping. Surface statistics: Precomputed mean and percentiles across 0.1° gridded (weathernext_3_0_0_0p1deg) and 0.05° station (weathernext_3_0_0_0p05deg) collections. Python (ee, geemap), JavaScript Earth Engine Guide
BigQuery SQL-based analytics, joining weather forecasts with enterprise business datasets (stores, assets, supply chain). Surface statistics: Precomputed mean and percentiles across 0.1° gridded (weathernext_3_0_0_0p1deg) and 0.05° station (weathernext_3_0_0_0p05deg) tables. SQL, BigQuery GIS, Looker, DataFrames BigQuery Guide

Step 3: Run your first query

Explore interactive Colab notebooks with end-to-end sample code and visualization tutorials:

Additional resources