Description of the upgrade

The IFS Cycle 49r1, will bring major changes to the IFS model and data assimilation system, including among many others:

• assimilation of 2m temperature observations, 
• Land-surface model updates,
• increased resolution and soft-recentering of the ensemble of data assimilations (EDA)  and
• activation of the stochastically perturbed parameterisations (SPP) model uncertainty scheme in all ensemble configurations. 

Extended-range forecast will change name to sub-seasonal range forecast. This name change will be gradually implemented in the documentation.

The re-forecast systems for medium and extended range will see the change in frequency, both will run over past 20 years, on the fixed days of the month: 

• medium range every four days: 1/5/9/13/17/21/25/29 (excluding 29 February)
• extended range every two days: 1/3/5/7/9/11/13/15/17/19/21/23/25/27/29/31 (excluding 29 February)

In IFS Cycle 49r1, HRES and the ENS control forecast become scientifically and computationally identical and both are run for 15-days at 00 UTC and 12 UTC, and 6 days at 06 UTC and 18 UTC. The superfluous ENS control forecast will be stopped in a future IFS upgrade, and the data stream currently known as HRES will become known as the “control” forecast.

From IFS Cycle 49r1 the wave model will be run on the same octahedral reduced gaussian grid as the atmospheric model, simplifying the data exchange between the two systems. Consequently the resolution of the wave model will increase to ~9km (O1280) in the medium range and ~36 km (O320) in the sub-seasonal range. With higher resolution, the wave model is better able to capture the atmospheric forcing, and details of the coastal conditions are generally better represented. The wave model will also run for 15-days at 00 UTC and 12 UTC, and for 6 days at 06 UTC and 18 UTC.

More detailed information will be made available in the coming weeks.

For any questions, please contact us via the ECMWF Support Portal .

#IFS49r1 #newfcsystem @ECMWF

News

Timeline of the implementation

Datasets affected

• HRES (day 1-15)

• ENS (day 1-15)

• ENS sub-seasonal (previously 'extended') (day 1-46)

• HRES-WAM (day 1-15)

• ENS-WAM (day 1-15)

• ENS-WAM sub-seasonal (previously 'extended') (day 1-46)

Resolution

Resolutions in bold increased/changed from the previous operational IFS cycle.

Dissemination schedule

The dissemination schedule will change for both the medium range and sub-seasonal range re-forecasts.

Medium range re-forecasts will be produced and disseminated every four days starting with 1st of every month:   1/5/9/13/17/21/25/29 (excluding 29 February)

Sub-seasonal range re-forecasts will be produced and disseminated every odd day of the month: 1/3/5/7/9/11/13/15/17/19/21/23/25/27/29/31 (excluding 29 February)

Meteorological content

Assimilation

• An upgrade of the Radiative Transfer for TOVS (RTTOV) model to version 13.2, which includes a new neural network-based surface emissivity model (SURFEM) trained on the reference Passive and Active Reference Microwave to Infrared Ocean (PARMIO) model. ​

• Non-microwave satellite observation package ​

• T2m assimilation package (read more in the article: Model upgrade will bring better two-metre temperature forecasts) ​

• VarQC and stratospheric balance assimilation package

• Activation of Variational Quality Control (VarQC) in the first 4D-Var minimization ​

• Land-surface assimilation package

• Assimilation of altimeter wave height data will be done in hourly sequential windows rather than 6‑hourly ones

• Updates to ensemble of data assimilations (EDA)

  • the horizontal grid-spacing of the EDA outer loop resolution is reduced to ~9 km and the inner-loop grid-spacing is reduced from ~100 km to ~40 km

  • soft-recentering of each member (1 outer loop) around a more accurate control member (3 outer loops)

Observations

• Several updates to non-microwave observations, including
  • reduced thinning of Spinning Enhanced Visible and InfraRed Imager (SEVIRI) data, and
  • scene dependent observation errors for Cross-track Infrared Sounders (CrIS, NPP and NOAA-20)
• The Global Navigation Satellite System Radio Occultation (GNSS-RO) assimilation is extended in the vertical from 50 km to 60 km in altitude

Model

• Wave model and convection package (read more in the article: Ocean wind wave model upgrade improves the forecasting system)

• Land-surface model updates (read more in the article: An urban scheme for the IFS)

• Activation of the Stochastically Perturbed Parametrisations (SPP) scheme, which replaces the effective but long-serving Stochastically Perturbed Parametrisation Tendencies (SPPT) scheme in all ensemble configurations

• Physics and numerics contributions

• Atmospheric composition package

Meteorological impact

Medium range

• 2m temperature and 10m wind speed forecasts are improved in short and medium-range.
  • The largest impacts on 2m temperature forecasts are for the winter months in the northern hemisphere, where the Continuous Ranked Probability Score (CRPS) is improved by 11% at day 1 and 2% at day 10.
  • Forecasts of 10 m wind speed are improved throughout the year. The largest impacts are for the winter months of the northern hemisphere, where the CRPS is improved by 12% at day 1 and 6% at day 10
• Increased spread in Tropical Cyclones – beneficial for strong systems.

Sub-seasonal range

• The most robust impacts on weekly mean forecast anomalies are small but statistically robust changes in the ensemble spread, which are driven by the switch from SPPT to SPP

• Positive impact on Madden-Julian Oscillation (MJO) at week 3-4

Evaluation

Interactive scorecards presenting the new cycle performance are coming soon.

Key configuration changes

Atmospheric model

• HRES and ENS Control run will be computationally identical.

• Both HRES and ENS will run:

  • to day 15 (step 360) for the 00 and 12 runs, and

  • to day 6 (step 144) for the 06 and 18 runs (as part of BC Optional Programme)

• There will be no change in vertical or horizontal resolution of the Atmospheric model 

• There will be no ENS configuration changes

• Medium range hindcast and Medium range hindcast statistics will be produced every four days starting from 1st of every month: 1/5/9/13/17/21/25/29 (excluding 29 February)
• Sub-seasonal hindcast and Sub-seasonal hindcast statistics will be produced every odd day of the month: 1/3/5/7/9/11/13/15/17/19/21/23/25/27/29/31 (excluding 29 February)

Wave model

• With IFS Cy49r1, the wave model will be run on the same grid as the atmospheric model, simplifying the data exchange between the two systems.  This means that the native grid for ENS-WAM will change from the 14km reduced lat-lon grid with 0.125° latitudinal spacing to the 9km O1280 octahedral reduced Gaussian grid.
• The wave spectrum on the O1280 grid continues to be discretised with 36 frequencies and 36 directions. However, to reduce the data volume, please note that in Cycle 49r1 only the first 29 frequencies will be output spanning a range from ~0.035 Hz to ~0.5 Hz.
• For users receiving parameter 2DFD via Dissemination, the number of frequencies will be adjusted automatically in the data requirements in the test system.  Users requesting 2DFD from MARS will need to adjust the number of frequencies in their retrieval requests.

• WAM and WAM sub-seasonal grid resolution will be the same as that of the corresponding ENS and ENS sub-seasonal. For the native grid, this means:
  • Medium range wave model will increase from a 14km reduced lat-lon grid with 0.125° latitudinal spacing to the 9km O1280 octahedral reduced Gaussian grid.
  • Sub-seasonal wave model will increase from a 55km reduced lat-lon grid with 0.125° latitudinal spacing to the 36km O320 octahedral reduced Gaussian grid.
• The number of frequencies output will be 29, spanning a range from ~0.035 Hz to ~0.5 Hz .
• For users receiving parameter 2DFD via Dissemination, this will be automatically adjusted in your data requirements in the test system.  Users requesting 2DFD from MARS will need to adjust the number of frequencies in their retrieval requests.

• The grid resolution of the medium range wave and sub-seasonal range wave hindcast and hindcast statistics will increase to match the corresponding medium and sub-seasonal forecast (Atmosphere)
• The production schedule of medium range wave and sub-seasonal range wave hindcast and hindcast statistics will match that of the medium range and sub-seasonal range hindcast and hindcast statistics of the Atmospheric model

New and changed parameters

New parameters

The table contains the list of parameters expected to be available with the new cycle implementation. They will be available as part of the test data. With implementation the parameters will be available in MARS and dissemination.  

All the new parameters will be available from the Medium range ensemble model.

Changes to existing parameters 

Simulated Satellite Data

* Spinning Enhanced Visible and Infrared Imager (SEVIRI)
** Flexible Combined Imager (FCI)
*** This parameter was added in between cycles IFS 48r1 and 49r1

For more details about the changes to the Simulated Satellite, including how to download the test data, data see Simulated Satellite Data update.

Other changes

• Probabilistic clear-air turbulence products will be extended to include contributions from non-orographic gravity-wave dissipation outside of convection areas.
• Sea surface temperature (SST) parameter values over land will change. The values over land are currently set to an unphysical, but spatially and temporally constant 273.16K. SST values over land will be set to that of the lake model's mixed layer temperature. The underlying field is much smoother, which helps give much more realistic values around coastlines when interpolating this field. 

Discontinued parameters

With 49r1, CAPE (parameter 59), will be discontinued and only the improved MUCAPE (parameter 228235) will be produced.

CIN (parameter 228001) parameter will be replaced by MUCIN (parameter 228236).

Total precipitation (parameter 228) will be removed from the analysis.