# WeatherBench 2 GenCast Ensemble Forecasts (2020)

## Facts

- Provider: Google DeepMind (WeatherBench 2)
- License: CC-BY-4.0 (https://storage.googleapis.com/weatherbench2/datasets/gencast/LICENSE)
- Format: Zarr (in sampled files)
- Stored at: Google Cloud Storage, gs://weatherbench2/datasets/gencast (us)
- Access: public, readable without an account
- DOI: https://doi.org/10.1038/s41586-024-08252-9
- Category: Climate & Weather
- Source: https://doi.org/10.1038/s41586-024-08252-9
- Page: https://scigantic.com/archive/17fb6421-ac36-5612-bd06-09afbd266162
- Keywords: Climate & Weather, weather forecasting, Google Cloud Storage

## About

Probabilistic 15-day global weather forecasts from GenCast, the diffusion-based ensemble weather model of Price et al. (Nature, 2024), published in the WeatherBench 2 bucket for benchmarking against IFS ENS and other models. Measured 2026-11-14: the 0.25 degree store (2020-1440x721.zarr) covers initialisation times every 12 hours from 2020-01-01 to 2020-12-31 (732 times), 56 ensemble members (sample dimension) and 30 lead times from 12 h to 360 h, at 500, 700 and 850 hPa plus surface fields. Variables include geopotential, temperature, specific humidity, u and v wind, wind speed, vertical velocity, 2 m temperature, 10 m wind and wind speed, mean sea level pressure, sea surface temperature and 12-hour and 24-hour total precipitation. The full ensemble is about 148 TB uncompressed. The folder also holds an ensemble-mean store, a _raw store, and conservative regrids at 240x121 (1.5 degree) and 64x32 with matching means, useful for CRPS and spread-skill evaluation. Consolidated Zarr; open with xarray.open_zarr and evaluate with the weatherbench2 package. Licensed CC BY 4.0 by Google LLC; cite Price et al. 2024.

## Load the data

```python
import gcsfs
fs = gcsfs.GCSFileSystem(token='anon')
fs.ls('weatherbench2/datasets/gencast/', detail=False)[:10]
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Data structure

File types in sampled files: Zarr

Example file: `datasets/gencast/2020-1440x721.zarr/10m_u_component_of_wind/.zarray`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/17fb6421-ac36-5612-bd06-09afbd266162
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/17fb6421-ac36-5612-bd06-09afbd266162
- Over MCP: `get_schema_card` with id `17fb6421-ac36-5612-bd06-09afbd266162` at https://api.scigantic.com/mcp

## Related topics

- [Climate & Weather datasets](https://scigantic.com/topics/climate-and-weather)
- [Weather forecasting datasets](https://scigantic.com/topics/weather-forecasting)
