AIFS Machine Learning data

ECMWF is now running its own Artificial Intelligence Forecasting System (AIFS) as part of its experiment suite. The AIFS consists of a deterministic model and an ensemble model. Further details about the deterministic model can be found at https://doi.org/10.48550/arXiv.2406.01465.

Use of the deterministic model output is governed by the Creative Commons CC-4.0-BY licence and the ECMWF Terms of Use. This means that the data may be redistributed and used commercially, subject to appropriate attribution. Output data from the ensemble model is not yet available.

How to access real-time open data

Real-time open data is available for the deterministic model only. For more information on accessing these products, please see user documentation or visit the ECMWF Support Portal.

The data are released 1 hour after the real-time dissemination schedule

Graphical products

Graphical products are available as part of the charts catalogue and the ecCharts application for both the deterministic and ensemble models. If the products are not found, please reload the page.

In both applications, search for “aifs” to find the relevant products.

Available Products

Deterministic model

The output of the AIFS deterministic experimental model is forecast with 6-hourly time steps out to 15 days initialised from the ECMWF operational analysis. Forecasts are produced four times per day (00/06/12/18UTC).

All forecasts are produced on an approximately 0.25 x 0.25-degree grid (N320 grid). Open data for the AIFS is disseminated on a regular 0.25 x 0.25-degree grid.

Upper-air fields are z (geopotential), q (specific humidity), t (temperature), u (U component of wind), v (V component of wind) and w (vertical velocity) on the following pressure levels: 1000hPa, 925hPa, 850hPa, 700hPa, 600hPa, 500hPa, 400hPa, 300hPa, 250hPa, 200hPa, 150hPa, 100hPa and 50hPa.

The single-level fields are msl (mean sea level pressure), sp (surface pressure), 10u (10 metre U wind component), 10v (10 metre U wind component), 2t (2 metre temperature), 2d (2 metre dewpoint temperature), tp (total precipitation, accumulated from the start of the forecast) and cp (convective precipitation, accumulated from the start of the forecast). Since 20th June 2024, Tropical Cyclone tracks are now available in BUFR4 format.

Ensemble model

The output of the AIFS probabilistic experimental model is forecast with 12-hourly time steps out to 15 days initialised from the ECMWF IFS ensemble initial conditions. Forecasts are produced two times per day (00/12UTC).

All forecasts are produced on an approximately 1 x 1-degree grid (O96 grid). Open data for the AIFS ensemble model is not yet available and will be disseminated soon.

Upper-air fields are z (geopotential), q (specific humidity), t (temperature), u (U component of wind), v (V component of wind) and w (vertical velocity) on the following pressure levels: 1000hPa, 925hPa, 850hPa, 700hPa, 600hPa, 500hPa, 400hPa, 300hPa, 250hPa, 200hPa, 150hPa, 100hPa and 50hPa.

The single-level fields are msl (mean sea level pressure), 10u (10 metre U wind component), 10v (10 metre U wind component), 2t (2 metre temperature) and tp (total precipitation, accumulated from the start of the forecast).

 

 

 

 

 

 

 

 

 

 

Period
From
Keywords
Provided by
ECMWF