Datasets
Forecasts using IFS EPS CY47R3b replicating operations (51 members, TCo639, ORCA025Z75). These forecasts were run after the HPCF move to Bologna on the ATOS system, and therefore are not necessarily bit-identical to the operational forecasts at the time - but are statistically equivalent. The ocean state has had a hydrostatically balanced estimate of anthropogenic influence (warming) since 1850-1900 removed. The CO2 concentration was set to 285 ppm.
Interval/period: N/A
Forecasts using IFS EPS CY47R3b replicating operations (51 members, TCo639, ORCA025Z75). These forecasts were run after the HPCF move to Bologna on the ATOS system, and therefore are not necessarily bit-identical to the operational forecasts at the time - but are statistically equivalent. These forecasts are part of a series of perturbed forecast experiments designed to explore and quantify human influence on exterme weather.
Interval/period: N/A
Forecasts using IFS EPS CY47R3b replicating operations (51 members, TCo639, ORCA025Z75). These forecasts were run after the HPCF move to Bologna on the ATOS system, and therefore are not necessarily bit-identical to the operational forecasts at the time - but are statistically equivalent. These forecasts are part of a series of perturbed forecast experiments designed to explore and quantify human influence on exterme weather.
Interval/period: N/A
A Baseline for Global Weather and Climate Simulations at 1 km Resolution, data is aggregated/truncated to 9km, and avail 3-hourly, above 300hPa only upper level U wind is correct, meridional winds and vorticity/divergence is corrupted. see the paper https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020MS002192
Examples
Interval/period: N/A
A Baseline for Global Weather and Climate Simulations at 1 km Resolution, data is aggregated/truncated to 9km, and avail 3-hourly. see the paper https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020MS002192
Examples
Interval/period: N/A
Comparison reference to 'A Baseline for Global Weather and Climate Simulations at 1 km Resolution', data is not aggregated and archived at 4.5km, avail 3-hourly. see also the paper https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020MS002192
Examples
Interval/period: N/A
A Baseline for Global Weather and Climate Simulations at 1 km Resolution, data is the reference at 9km with convection parametrisation, NDJF avail 3-hourly. see the paper https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020MS002192
Examples
Interval/period: N/A
Forecasts using IFS EPS CY47R2 replicating operations (TCo639, ORCA025Z75). In this experiment, 251 members are forecast, rather than the usual 51. If you are interested in using these data, please let us know by contacting nicholas.leach@physics.ox.ac.uk .
Examples
Interval/period: N/A
Forecasts using IFS EPS CY47R2 replicating operations (TCo639, ORCA025Z75). In this experiment, 251 members are forecast, rather than the usual 51. The ocean state has had a hydrostatically balanced estimate of anthropogenic influence (warming) since 1850-1900 added, such that the total anthropogenic influence is twice estimated at the present day (an equal but opposite perturbation to expid b2mg). The CO2 concentration was set to 615 ppm.
Interval/period: N/A
Forecasts using IFS EPS CY47R2 replicating operations (TCo639, ORCA025Z75). In this experiment, 251 members are forecast, rather than the usual 51. The ocean state has had a hydrostatically balanced estimate of anthropogenic influence (warming) since 1850-1900 removed. The CO2 concentration was set to 285 ppm. If you are interested in using these data, please let us know by contacting nicholas.leach@physics.ox.ac.uk .
Examples
Interval/period: N/A
101-member coupled IFS (cycle 47R3) subseasonal reforecasts covering the period 2001-2020 with forecasts initialized on February 1st, May 1st, August 1st, and November 1st. The atmosphere is configured with 137 vertical levels and uses the Tco319 cubic octahedral reduced Gaussian grid. The IFS is coupled hourly to the 75 level version of the NEMO v3.4 ocean model and the LIM2 sea-ice model, both of which use the ORCA025 tripolar grid. Coupling follows the implementation used in ECMWF CY47R3 operational reforecasts.
Interval/period: N/A
Single-member eddy-permitting reforecasts based on the IFS cycle 48r1 atmosphere run at Tco319 combined with the eORCA025 NEMO4-SI3 ocean and sea ice configurations. Forecasts consists of a single unperturbed member run for 32 days starting on the first of each month between 1995 and 2016, for a total of 264 start dates. Atmospheric fields are initialised from the the GLORYS12v1 eddy-rich global ocean reanalysis.
Examples
Interval/period: N/A
Single-member eddy-rich reforecasts based on the IFS cycle 48r1 atmosphere run at Tco319 combined with the eORCA12 NEMO4-SI3 ocean and sea ice configurations. Forecasts consists of a single unperturbed member run for 32 days starting on the first of each month between 1995 and 2016, for a total of 264 start dates. Atmospheric fields are initialised from the the GLORYS12v1 eddy-rich global ocean reanalysis.
Examples
Interval/period: N/A
Updated 2025-11-04. A 10+1 member coupled IFS (cycle 47R3) extended-range reforecast experiment covering the period 1959-2023 with forecasts initialized on the 1st, 8th, 15th, and 22nd, of each month. The atmosphere is configured with 137 vertical levels and uses the Tco199 cubic octahedral reduced Gaussian grid. The IFS is coupled hourly to the 75 level version of the NEMO v3.4 ocean model and the LIM2 sea-ice model, both of which use the ORCA1 tripolar grid. Coupling follows the implementation used in ECMWF CY47R3 operational reforecasts.
Interval/period: N/A
101-member coupled IFS (cycle 47R3) subseasonal reforecasts covering the period 2001-2020 with forecasts initialized on February 1st, May 1st, August 1st, and November 1st. The atmosphere is configured with 137 vertical levels and uses the Tco199 cubic octahedral reduced Gaussian grid. The IFS is coupled hourly to the 75 level version of the NEMO v3.4 ocean model and the LIM2 sea-ice model, both of which use the ORCA1 tripolar grid. Coupling follows the implementation used in ECMWF CY47R3 operational reforecasts.
Interval/period: N/A
Single-member eddy-permitting reforecasts based on the IFS cycle 48r1 atmosphere run at Tco1279 combined with the eORCA025 NEMO4-SI3 ocean and sea ice configurations. Forecasts consists of a single unperturbed member run for 32 days starting on the first of each month between 1995 and 2016, for a total of 264 start dates. Atmospheric fields are initialised from the the GLORYS12v1 eddy-rich global ocean reanalysis.
Examples
Interval/period: N/A
Single-member eddy-rich reforecasts based on the IFS cycle 48r1 atmosphere run at Tco1279 combined with the eORCA12 NEMO4-SI3 ocean and sea ice configurations. Forecasts consists of a single unperturbed member run for 32 days starting on the first of each month between 1995 and 2016, for a total of 264 start dates. Atmospheric fields are initialised from the the GLORYS12v1 eddy-rich global ocean reanalysis.
Examples
Interval/period: N/A
Case study with an experimental IFS EDT configuration (CY49R2). Initialized on 2024-05-27 at 00Z. Outputting votemper, vosaline, vozocrtx, vomecrty on all ocean model levels.
Examples
retrieve, class=rd, date=2024-05-27, expver=inq9, stream=oper, type=fc, levtype=sfc, param=2t, time=00:00:00, step=24, target='output.grib'2m-temperature at step 24
Interval/period: N/A
nextGEMS cycle2 coupled global 1-year simulation without parameterized deep convection. Grids: tco2559 for atmosphere and NG5 for ocean. Model versions: IFS 47r3.3 + upgrades and FESOM2.1. Starting date: 20/01/2020.
Examples
retrieve, class=rd, date=20200120, stream=LWDA, expver=hqys, type=fc, levtype=sfc, param=2t, anoffset=9, time=0, step=24, target='output.grib'Retrieve 2m temperature after 24 hours
Interval/period: N/A
nextGEMS cycle2 coupled global 1-year simulation without parameterized deep convection. Grids: tco1279 for atmosphere and orca025 for ocean. Model versions: IFS 47r3.3 + upgrades and NEMO3.4. Starting date: 20/01/2020.
Examples
retrieve, class=rd, date=20200120, stream=LWDA, expver=hr2n, type=fc, levtype=sfc, param=2t, anoffset=9, time=0, step=24, target='output.grib'Retrieve 2m temperature after 24 hours
Interval/period: N/A
nextGEMS cycle2 coupled global 2-year simulation with parameterized deep convection. Grids: tco1279 for atmosphere and orca025 for ocean. Model versions: IFS 47r3.3 + upgrades and NEMO3.4. Starting date: 20/01/2020.
Examples
retrieve, class=rd, date=20200120, stream=LWDA, expver=hrn6, type=fc, levtype=sfc, param=2t, anoffset=9, time=0, step=24, target='output.grib'Retrieve 2m temperature after 24 hours
Interval/period: N/A
IFS-COMPO CY49R1 free-running simulation of atmospheric composition (chemistry, aerosol), with meteorology re-initialized from ERA5 every day at 00UTC, with three-hourly output for the years 2003 to 2005. Spinup-time is two months (Nov-Dec 2002) and grid is reduced linear gaussian, TL255 (~80km horizontal resolution). Instantaneous aerosol MMR fields are stored on 16 pressure levels.
Examples
Interval/period: N/A
IFS-COMPO CY49R1 free-running simulation of atmospheric composition (chemistry, aerosol), with meteorology re-initialized from ERA5 every day at 00UTC, with three-hourly output for the years 2006 to 2008. Spinup-time is two months (Nov-Dec 2005) and grid is reduced linear gaussian, TL255 (~80km horizontal resolution). Instantaneous aerosol MMR fields are stored on 16 pressure levels.
Examples
Interval/period: N/A
IFS-COMPO CY49R1 free-running simulation of atmospheric composition (chemistry, aerosol), with meteorology re-initialized from ERA5 every day at 00UTC, with three-hourly output for the years 2009 to 2011. Spinup-time is two months (Nov-Dec 2008) and grid is reduced linear gaussian, TL255 (~80km horizontal resolution). Instantaneous aerosol MMR fields are stored on 16 pressure levels.
Examples
Interval/period: N/A
IFS-COMPO CY49R1 free-running simulation of atmospheric composition (chemistry, aerosol), with meteorology re-initialized from ERA5 every day at 00UTC, with three-hourly output for the years 2012 to 2014. Spinup-time is two months (Nov-Dec 2011) and grid is reduced linear gaussian, TL255 (~80km horizontal resolution). Instantaneous aerosol MMR fields are stored on 16 pressure levels.
Examples
Interval/period: N/A