# Overview

This dataset contains sub-seasonal forecasts submitted by participants in the AI Weather Quest, a global AI/ML competition organised by ECMWF to advance sub-seasonal forecasting. Please note the forecasts reflect participants' model outputs and are provided as-is to promote transparency and reproducibility in AI-based weather prediction.

More information on the competition is available here: https://aiweatherquest.ecmwf.int/

In addition to this README file, supporting documentation is available here:
https://confluence.ecmwf.int/display/AWQ/Detailed+guide+for+downloading+forecast+data

## Submitted forecasts overview

Teams were challenged to submit global, quintile-based probabilistic forecasts of at least one of the following variables:

- Weekly-mean near-surface (2m) temperature (tas)
- Weekly-mean sea level pressure (mslp)
- Weekly-accumulated precipitation (pr)

Forecasts were requested at a 1.5-degree latitude/longitude resolution for one of the following lead times (inclusive):

- Days 19 to 25 (week 3)
- Days 26 to 32 (week 4)

## Filename formatting

A separate NetCDF file is provided for each forecasted variable submitted by a team. Files are named using the following naming convention:
<<variable>>_<<fc_start_date>>_p<<fc_window>>_<<teamname>>_<<modelname>>.nc 
Where - 

variable (str): The forecasted variable. Options are:
'tas': Near-surface temperature (weekly-mean)
'mslp': Mean sea level pressure (weekly_mean)
'pr': Precipitation (weekly-accumulated)

fc_start_date (str): The forecast initialisation date in format YYYYMMDD (e.g., ‘20250515’ for 15th May 2025).

fc_window (str or int): The selected forecasting window. Valid options are:
'1': Weekly-mean forecasts for days 19 to 25 inclusive.
'2': Weekly-mean forecasts for days 26 to 32 inclusive.

teamname (str): The team name associated with the submitted forecast.

modelname (str): The model name associated with the submitted forecast.

Information regarding individual teams and models can be found at: https://aiweatherquest.ecmwf.int/teams/

## Directory structure

To facilitate access, forecasts are organised into directories that can be browsed by either:

1) Forecast initialisation date (by_fc_date)
	- by_fc_date -> by_team -> by_model -> forecasts

2) Participating team (by_team)
	- by_team -> by_model -> by_fc_date -> forecasts

For more efficient 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)

Further guidance on file content and recommended download methods is available on the following confluence page: https://confluence.ecmwf.int/display/AWQ/Detailed+guide+for+downloading+submitted+forecast+data

## License and Dataset Citation

This dataset is released under ECMWF's general licence and the Creative Commons Attribution 4.0 International (CC BY 4.0) licence. The following provide full licence details: https://apps.ecmwf.int/datasets/licences/general & https://creativecommons.org/licenses/by/4.0/

Users should cite the dataset DOI (provided on https://www.ecmwf.int/en/forecasts/dataset/ai-weather-quest-sub-seasonal-forecasts) and the following reference when using this dataset:

Loegel, O., Talib, J., Vitart, F., Hoffmann, J. and Chantry, M., 2025. The AI Weather Quest: an international competition for sub-seasonal forecasting with AI. Machine Learning: Earth, 1(1), p.010701. https://doi.org/10.1088/3049-4753/adf649

If applicable, please acknowledge the AI Weather Quest competition as follows: “This dataset was produced as part of the AI Weather Quest competition organised by ECMWF (2025-2026)".