...
All forecasts submitted to the AI Weather Quest are openly available via the AI Weather Quest Data Portal. Participants submit quintile-based probabilistic forecasts on a 1.5 degree latitude/longitude grid for near-surface air temperature (tas), mean sea level pressure (mslp), and accumulated precipitation (pr), targeting lead times of days 19 to 25 (week 3) and 26 to 32 (week 4). This guide explains how to download, extract, and use individual forecast files or archives.
...
The following directory structures are followed:
- by_fc_date >> / -> by_team >> / -> by_model >> / -> forecasts
- by_team >> / -> by_model >> / -> by_fc_date >> forecasts/ -> forecasts
Each forecast A separate NetCDF file is provided for each forecasted variable and lead time submitted by a team. is stored as a separate NetCDF (.nc) file.
Zip/Tar Archives
For more efficient simplify downloading, we also provide .tar.gz archives containing all forecasts either:
- submitted by a given team (by_team_zipfiles), or
- issued for a given initialisation date (by_fc_date_zipfiles)
...
tar -czf AIFS.tar.gz AIFS_forecasts/
Forecast file content
All forecasts are saved in NetCDF files created using the AI_WQ_create_empty_dataarray function in the forecast_submission module of AI Weather Quest Python Package.
All data is stored in a fractional format.
Dimensions
| Dimension name | Characteristics |
|---|---|
| Quintile | Labelled [0.2,0.4,0.6,0.8,1.0], where the quintile value represents the upper limit of climatological conditions. |
| Latitude | At a 1.5 degree spacing from 90.0 to -90.0. |
| Longitude | At a 1.5 degree spacing from 0.0 to 360.0. |
The files include additional attributes to support file storage and forecast visualisation. Attributes important to forecast users include:
| Attribute | Description |
|---|---|
| Forecast_issue_date | Forecast initialisation date (np.datetime64). |
| Forecast_period_start | Start of forecasting window (np.datetime64). Aids distinction between forecasting windows. |
| Forecast_period_end | End of forecasting window (np.datetime64). Please note, for tas and mslp ends at 18:00:00 whilst pr ends at 00:00:00 due to differences between weekly-means and weekly-accumulations. |
| teamname | Team name associated with the submitted forecast. |
| modelname | Model name associated with the submitted forecast. |
Examples of opening forecasts
Forecasts are easy to open in a Python environment. We highly recommend using xarray to load in an individual forecast:
import xarray as xr
fc = xr.load_dataarray([filename])
We also recommend using xarray to open multiple forecasts of the same variable and forecasting window from different models. As an example:
fcs = xr.load_dataarray('tas_20250814_p1_*.nc',combine='nested',concat_dim='modelname')This output has an additionally dimension called modelname.
Licence information
This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence. Full licence terms can viewed here.
...