Please provide a short description of the Ensemble Prediction System
Global ensemble system that simulates initial
unceratinties
uncertainties using singular vectors and perturbations from an ensemble of data assimiliations. Model uncertainties represented with SPPT and SKEB. Based on 51 members, run twice-a-day up to day 15, (extended to 32 days twice weekly, on Mondays and Thursdays).
Global ensemble system that simulates initial uncertainties using singular vectors and model uncertainties due to physical parameterisations using a stochastic scheme. Based on 51 members, run twice-a-day up to day 15, with at 00UTC a coupled ocean system from day 10 to day 15 (extended to 32 days once a week, on Thursdays).
Research or Operational? If not operational, are there any plans to become so?
Operational
Operational
Global or Regional EPS? (See section 7 for items specific to regional EPS)
Global
Data time of first forecast run
Date of last forecast with this version (if applicable)
Data time of last forecast run (if applicable)
Is there a higher-resolution control forecast available? (If yes, this should be described in a separate sheet of this spreadsheet.)
No
No
Brief summary of main changes from previous version (keywords).
Coupling to NEMO ocean model, 91 levels with top a 1 Pa, higher horizontal resolution, EDA initial perturbations, revised model uncertainty representation
N/A - First version listed
2. Configuration of the EPS
Horizontal resolution of the model. (Where variable resolution is used, please describe in full.)
TL639
TL399
Horizontal configuration and resolution of the output grid
TL639 L91 for day 1 to day 10 (leg 1) and TL319 L91 after day 10 (leg 2) The resolution archived is N320 reduced gaussian grid for leg1 and N160 reduced gaussian grid for leg2.
T399 L62 for day 1 to day 10 (leg 1) and T255 L62 for T+246 to day 15 (leg 2) The resolution archived is N200 reduced gaussian grid for leg1 and N128 reduced gaussian grid for leg2.
Number of model levels
91
62
Type of model levels (eg sigma)
sigma
sigma
Forecast length and forecast step interval
T+0h to T+360h at 6h
T+0h to T+360h at 6h
Runs per day (Times in UTC)
2 (00, 12)
2 (00, 12)
Is there an unperturbed control forecast included? (Y/N)
Y
Y
Number of perturbed ensemble members (excluding control)
50
50
Integration time step
20 min for leg 1 and 45 min for leg 2
30 min
Top of model - model section
~0.01hPa
~5hPa
Is model coupled to an ocean model?
Yes
No
If yes, please describe ocean model briefly including any ensemble perturbations applied
NEMO 1deg, 5 different ocean analyses
Additional comments
3. Initial conditions and Perturbations
Data assimilation method for control analysis
4D-Var 12h window
4D-Var 12h window
Resolution of model used to generate control analysis
TL1279L137
TL799L91
Control variables used in data assimilation
N/A
N/A
Ensemble initial perturbation strategy
Singular Vectors (Total energy norm) and ensemble of data assimilations (EDA)
Singular Vectors (Total energy norm)
Optimisation time in forecast (if applicable)
T+48
T+48
Horizontal resolution of perturbations (if different from model resolution)
singular vectors T42L91 and TL399L137 for EDA
T42L62
Initial perturbed area
global
Extra tropical (<30S, >30N) + up to 6 tropical areas
Are perturbations to observations employed? (Y/N)
YY
Yes
No
Perturbations added to control analysis or derived directly from ensemble analysis
Added
Added
Perturbations in +/- pairs? (Y/N)
Y
Yes
Additional comments
N/A
N/A
4. Model Uncertainty Perturbations
Is model physics perturbed? If yes, briefly describe method(s).
Y. Uses Stochastically Perturbed Parameterization Tendencies (SPPT) and Stochastic Kinetic Energy Backscatter (SKEB)
Y. Stochastic perturbation of physics tendency by factor in range [0.5,1.5]
Do all ensemble members use exactly the same model version, or are, for example, different
parameterization
parameterisation schemes used? Please describe any differences.
Same
Same
Is model dynamics perturbed? If yes, briefly describe method(s).
N
No
N
Are the above model uncertainty perturbations applied to the control forecast?
No
Additional comments
N/A
N
Additional comments
/A
5. Surface Boundary Perturbations
Perturbations to sea-surface temperature? If yes, briefly describe method(s).
Y
Yes, 5 different ocean analyses
N
No
Perturbations to soil moisture? If yes, briefly describe method(s).
Y
Yes, from EDA
NN
No
Perturbations to surface wind stress or roughness? If yes, briefly describe method(s).
N
N
No
Any other surface perturbations? If yes, briefly describe method(s).
N
No
Are the above surface perturbations applied to the control forecast?
Data policy of originating centre for usage of data in TIGGE
Users of the ECMWF data sets are requested to reference the source of the data in any publication, e.g. "ECMWF ERA-40 data used in this study/project have been provided by ECMWF/have been obtained from the ECMWF Data Server".
Users of the ECMWF data sets are requested to reference the source of the data in any publication, e.g. "ECMWF ERA-40 data used in this study/project have been provided by ECMWF/have been obtained from the ECMWF Data Server".
9. TIGGE Specific Information
Version Identifier Code
Date of first forecast in TIGGE
1st October 2006
1st October 2006
Data time of first forecast run in TIGGE
00Z
0 Z
00Z
Date of last forecast in TIGGE
N/A
N/A
Data time of last forecast run in TIGGE
N/A
N/A
Is there a higher-resolution control forecast included in TIGGE? If so give tab name where it is described.
Yes, there is a control forecast run at TL639 and a high resolution forecast run at TL1279
Yes, there is a control forecast run at T399 and a high resolution forecast run at T799
Brief summary of main changes from previous version (keywords).
Resolution, coupling to ocean, representation of initial and model uncertainties
N/A - First version in TIGGE
Key reference papers for EPS
...
Buizza, R., & Palmer, T. N., 1995: The singular-vector structure of the atmospheric general circulation. J. Atmos. Sci., 52, 9, 1434-1456.
...
Molteni,F., Buizza,R., Palmer,T.N. and Petroliagis,T., 1996: The ECMWF Ensemble Prediction System: Methodology and Validation Q.J.R Meteorol.Soc. (1996) Vol 122, pp 73-119.
...
Buizza, R., Miller, M., & Palmer, T. N., 1999a: Stochastic representation of model uncertainties in the ECMWF Ensemble Prediction System. Q. J. R. Meteorol. Soc., 125, 2887-2908.
...
Buizza, R., Bidlot, J.-R., Wedi, N., Fuentes, M., Hamrud, M., Holt, G., & Vitart, F., 2007: The new ECMWF VAREPS (Variable Resolution Ensemble Prediction System). Q. J. Roy. Meteorol. Soc., 133, 681-695.