Table of Contents |
---|
1. Ensemble version | GloSea6 | GloSea5 |
---|---|---|
Ensemble identifier code | HadGEM3 GC3.2 | HadGEM3 GC2.0 |
Short Description | Global ensemble system that simulates initial-condition uncertainties using lagged initialisation and model uncertainties using a stochastic scheme. There are 4 ensemble members initialised each day, each extending to 60 days. | Global ensemble system that simulates initial-condition uncertainties using lagged initialisation and model uncertainties using a stochastic scheme. There are 4 ensemble members initialised each day, each extending to 60 days. |
Research or operational | Operational | Operational |
Data time of first forecast run | 02/02/2021 | 05/02/2015 |
2. Configuration of the EPS | ||
Is the model coupled to an ocean model? | Yes from day 0 | Yes from day 0 |
If yes, please describe ocean model briefly including frequency of coupling and any ensemble perturbation applied | Ocean model is Global Ocean 6.0, based on NEMO3.6 with 0.25 degree horizontal resolution, 75 vertical levels, initialized using NEMOVAR; no perturbations. Frequency of coupling is 1-hourly. | Ocean model is Global Ocean 6.0, based on NEMO3.6 with 0.25 degree horizontal resolution, 75 vertical levels, initialized using NEMOVAR; no perturbations. Frequency of coupling is 1-hourly. |
If no, please describe the sea surface temperature boundary conditions (climatology, reanalysis ...) | ||
Is the model coupled to a sea ice model? | Yes | Yes |
If yes, please describe sea-ice model briefly including any ensemble perturbation applied | Global Sea Ice 8.1 (CICE5.1.2) initialized from NEMOVAR; no perturbations. | Global Sea Ice 8.1 (CICE5.1.2) initialized from NEMOVAR; no perturbations. |
Is the model coupled to a wave model? | No | No |
If yes, please describe wave model briefly including any ensemble perturbation applied | ||
Ocean model | NEMO 0.25 degree resolution | NEMO 0.25 degree resolution |
Horizontal resolution of the atmospheric model | N216 0.83° x 0.56° (approx 60km in mid-latitudes) | N216 0.83° x 0.56° (approx 60km in mid-latitudes) |
Number of model levels | 85 | 85 |
Top of model | 85 km | 85 km |
Type of model levels | terrain-following, height-based vertical coordinate | terrain-following, height-based vertical coordinate |
Forecast length | 60 days | 60 days |
Run Frequency | daily | daily |
Is there an unperturbed control forecast included? | No | No |
Number of perturbed ensemble members | 4 per day | 4 per day |
Integration time step | 15 minutes | 15 minutes |
3. Initial conditions and perturbations | ||
Data assimilation method for control analysis | 4D Var | 4D Var |
Resolution of model used to generate Control Analysis | N1280L70 (0.23° x0.16°) | N768L70 (0.23° x 0.16°) |
Ensemble initial perturbation strategy | lagged initialisation | lagged initialisation |
Horizontal and vertical resolution of perturbations | N/A | N/A |
Perturbations in +/- pairs | N/A | N/A |
Additional comments | Soil moisture is initialised using Met Office JULES-JRA55 analysis. | Soil moisture is initialised with climatological mean values in both real-time forecasts and re-forecasts. |
Initialization of land surface | ||
3.1 What is the land surface model (LSM) and version used in the forecast model, and what are the current/relevant references for the model? Are there any significant changes/deviations in the operational version of the LSM from the documentation of the LSM? | The Met Office Seasonal Forecast System version 6 using Global Coupled 3.2 (GloSea6-GC3.2) uses the Joint UK Land Environment Simulator (JULES). | The Met Office Seasonal Forecast System version 5 using Global Coupled 2.0 (GloSea5-GC2) uses the Joint UK Land Environment Simulator (JULES). |
3.2 How is soil moisture initialized in the forecasts? (climatology / realistic / other)? If “climatology”, what is the source of the climatology? If “realistic”, does the soil moisture come from an analysis using the same LSM as is coupled to the GCM for forecasts, or another source? Please describe the process of soil moisture initialization. If “other”, please describe the process of soil moisture initialization. | In GloSea6-GC3.2 the soil moisture is initialised from a JULES analysis forced with JRA-55 analysis. | In GloSea5-GC2 the soil moisture is initialised from a seasonally varying climatology. This climatology was derived from a JULES re-analysis using Global Land 3.0 and forced with the WATCH-Forcing-Data-ERA-Interim forcing set (Wheedon et al, 2014). This re-analysis was completed on a 0.5 degree grid and interpolated to the model resolution (0.83 x 0.56 degrees). The climatology from this re-analysis has been scaled to match the climatology of our NWP soil moisture climatology. |
3.3 How is snow initialized in the forecasts? | ||
If “climatology”, what is the source of the climatology? | ||
If “realistic”, does the snow come from an analysis using the same LSM as is coupled to the GCM for forecasts, or another source? Please describe the process of soil moisture initialization. | ||
If “other”, please describe the process of soil moisture initialization. | In GloSea6-GC3.2 the soil moisture is initialised from a JULES analysis forced with JRA-55 analysis. | |
Is there horizontal and/or vertical interpolation of data onto the forecast model grid? | ||
Are snow mass, snow depth or both initialized? | Snow is initialised “realistically” from analysis using JRA-55. | Snow is initialised “realistically” from analysis. For the hindcasts this is ERA-Interim and the forecasts use the Met Office NWP global analysis. The Met Office NWP global model uses the same land surface model as GloSea5-GC2. For the hindcast the snow field is interpolated from 0.75x0.75 degrees (ERA-I) to the GloSea5-GC2 grid. Only snow mass is initialized. |
3.4 How is soil temperature initialized in the forecasts? (climatology / realistic / other) | ||
If “climatology”, what is the source of the climatology? | ||
If “realistic”, does the soil moisture come from an analysis using the same LSM as is coupled to the GCM for forecasts, or another source? Please describe the process of soil moisture initialization. | ||
If “other”, please describe the process of soil moisture initialization. | ||
Is the soil temperature initialized consistently with soil moisture (frozen soil water where soil temperature ≤0°C) and snow cover (top layer soil temperature ≤0°C under snow)? | ||
Is there horizontal and/or vertical interpolation of data onto the forecast model grid? | ||
If all model soil layers are not initialized in the same way or from the same source, please describe. | ||
If all model soil layers are not initialized in the same way or from the same source, please describe. | Soil temperature is initialised “realistically” from analysis using JRA-55. | Soil temperature is initialised “realistically” from analysis. For the hindcasts this is ERA-Interim and the forecasts use the Met Office NWP global analysis. For the hindcast the soil temperature field is interpolated from 0.75x0.75 degrees (ERA-I) to the GloSea5-GC2 grid. The level in the ERA-interim LSM start at 0, 7, 28, 100cm (https://confluencesoftware.ecmwf.int/wiki/pages/viewpage.action?pageId=56660259). The GloSea5-GC2 soil model levels are (in metres): (0.0,0.10), (0.10,0.35), (0.35,1.0), (1.0,3.0) |
3.5 How are time-varying vegetation properties represented in the LSM? | ||
Is phenology predicted by the LSM? If so, how is it initialized? | ||
If not, what is the source of vegetation parameters used by the LSM? Which time-varying vegetation parameters are specified (e.g., LAI, greenness, vegetation cover fraction) and how (e.g., near-real-time satellite observations? Mean annual cycle climatology? Monthly, weekly or other interval?) | We do not include phenology. GloSea6 uses a fraction tile system with 9 tiles: 5 plant functional types and 4 non-vegetated types. The fractional values are derived from IGBP. Canopy height of plant functional types is derived from MODIS LAI data. The following variable is time varying and derived from MODIS LAI data: * Leaf area index of plant functional types This variable is specified at monthly intervals but there is no inter-annual variation. The initialisation values are interpolated from the monthly time series. | We do not include phenology. GloSea5 uses a fraction tile system with 9 tiles: 5 plant functional types and 4 non-vegetated types. The fractional values are derived from IGBP. Canopy height of plant functional types is derived from MODIS LAI data. The following variable is time varying and derived from MODIS LAI data: * Leaf area index of plant functional types This variable is specified at monthly intervals but there is no inter-annual variation. The initialisation values are interpolated from the monthly time series. |
3.6 What is the source of soil properties (texture, porosity, conductivity, etc.) used by the LSM? | The soil information is derived from the Harmonized World Soil Database. | The soil information is derived from the Harmonized World Soil Database. |
3.7 If the initialization of the LSM for re-forecasts deviates from the procedure for forecasts, please describe the differences. | The soil moisture, temperature and snow in the re-forecasts are taken from a JULES reanalysis forced by the JRA-55 reanalysis. Shinya KOBAYASHI, Yukinari OTA, Yayoi HARADA, Ayataka EBITA, Masami MORIYA, Hirokatsu ONODA, Kazutoshi ONOGI, Hirotaka KAMAHORI, Chiaki KOBAYASHI, Hirokazu ENDO, Kengo MIYAOKA, Kiyotoshi TAKAHASHI (2015), The JRA-55 Reanalysis: General Specifications and Basic Characteristics, Journal of the Meteorological Society of Japan. Ser II. https://doi.org/10.2151/jmsj.2015-001 | There are differences between the forecast and re-forecast initialisation. These are described in the relevant sections. |
4. Model uncertainties perturbations | ||
Is model physics perturbed? | Yes. A scheme called Stochastic Kinetic Energy Backscatter scheme (SKEB) adds vorticity perturbations to the forecast in order to counteract the damping of small-scale features introduced by the semi-Lagrangian advection scheme. | Yes. A scheme called Stochastic Kinetic Energy Backscatter scheme (SKEB) adds vorticity perturbations to the forecast in order to counteract the damping of small-scale features introduced by the semi-Lagrangian advection scheme. |
Do all ensemble members use exactly the same model version? | Yes | Yes |
Is model dynamics perturbed? | No | No |
Are the above model perturbations applied to the control forecast? | Yes | Yes |
5. Surface boundary perturbations | ||
Perturbations to sea surface temperature? | No | No |
Perturbation to soil moisture? | No | No |
Perturbation to surface stress or roughness? | No | No |
Any other surface perturbation? | No | No |
Are the above surface perturbations applied to the Control forecast? | N/A | N/A |
Additional comments | As the perturbation are exclusively based on stochastic physics and are applied to all forecast members, there is no true control member. | As the perturbation are exclusively based on stochastic physics and are applied to all forecast members, there is no true control member. |
6. Other details of the models | ||
Description of model grid | Arakawa-C | Arakawa-C |
List of model levels in appropriate coordinates | Level list (km) 0.0200000, 0.0533333, 0.100000, 0.160000, 0.233333, 0.320000, 0.420000, 0.533333, 0.660000, 0.800000, 0.953334, 1.12000, 1.30000, 1.49333, 1.70000, 1.92000, 2.15333, 2.40000, 2.66000, 2.93333, 3.22000, 3.52000, 3.83333, 4.16000, 4.50000, 4.85333, 5.22000, 5.60000, 5.99333, 6.40000, 6.82000, 7.25333, 7.70000, 8.16000, 8.63334, 9.12001, 9.62002, 10.1334, 10.6601, 11.2002, 11.7536, 12.3205, 12.9009, 13.4949, 14.1025, 14.7239, 15.3592, 16.0088, 16.6729, 17.3519, 18.0463, 18.7567, 19.4839, 20.2288, 20.9925, 21.7765, 22.5824, 23.4122, 24.2682, 25.1532, 26.0706, 27.0241, 28.0183, 29.0582, 30.1500, 31.3005, 32.5177, 33.8106, 35.1895, 36.6662, 38.2540, 39.9679, 41.8249, 43.8438, 46.0462, 48.4558, 51.0994, 54.0064, 57.2100, 60.7467, 64.6570, 68.9855, 73.7818, 79.1000, 85.0000 | Level list (km) 0.0200000, 0.0533333, 0.100000, 0.160000, 0.233333, 0.320000, 0.420000, 0.533333, 0.660000, 0.800000, 0.953334, 1.12000, 1.30000, 1.49333, 1.70000, 1.92000, 2.15333, 2.40000, 2.66000, 2.93333, 3.22000, 3.52000, 3.83333, 4.16000, 4.50000, 4.85333, 5.22000, 5.60000, 5.99333, 6.40000, 6.82000, 7.25333, 7.70000, 8.16000, 8.63334, 9.12001, 9.62002, 10.1334, 10.6601, 11.2002, 11.7536, 12.3205, 12.9009, 13.4949, 14.1025, 14.7239, 15.3592, 16.0088, 16.6729, 17.3519, 18.0463, 18.7567, 19.4839, 20.2288, 20.9925, 21.7765, 22.5824, 23.4122, 24.2682, 25.1532, 26.0706, 27.0241, 28.0183, 29.0582, 30.1500, 31.3005, 32.5177, 33.8106, 35.1895, 36.6662, 38.2540, 39.9679, 41.8249, 43.8438, 46.0462, 48.4558, 51.0994, 54.0064, 57.2100, 60.7467, 64.6570, 68.9855, 73.7818, 79.1000, 85.0000 |
What kind of large scale dynamics is used? | Semi-lagrangian | Semi-lagrangian |
What kind of boundary layer parameterization is used? | Nolocal mixing scheme and local Richardson number scheme | Nolocal mixing scheme and local Richardson number scheme |
What kind of convective parameterization is used? | Mass flux scheme | |
What kind of large-scale precipitation scheme is used? | Walters et al 2017 | Williams et al., 2015 |
What cloud scheme is used? | Prognostic cloud fraction | Prognostic cloud fraction |
What kind of land-surface scheme is used? | Jules coupled model; Best et al 2011 | Jules coupled model; Best et al 2011 |
How is radiation parametrized? | Walters et al 2017 | Williams et al 2015 |
Other relevant details? | ||
7. Re-forecast configuration | ||
Number of years covered | 23 years (1993-2015) | 23 years (1993-2015) |
Produced on the fly or fix re-forecasts? | On the fly | On the fly |
Frequency | each month, on 1st, 9th, 17th, 25th | each month, on 1st, 9th, 17th, 25th |
Ensemble size | 7 members per year (from 25 March 2017 hindcast onwards, prior to this 3 members per year) | 7 members per year (from 25 March 2017 hindcast onwards, prior to this 3 members per year) |
Initial conditions | ERA interim and NEMOVAR | ERA interim and NEMOVAR |
Is the model physics and resolution the same as for the real-time forecasts? | Yes | Yes |
If not, what are the differences | N/A | N/A |
Is the ensemble generation the same as for real-time forecasts? | Yes | Yes |
If not, what are the differences | N/A | N/A |
8. References
- Bowler N, Arribas A, Beare S, Mylne KE, Shutts G. 2009. The local ETKF and SKEB: Upgrades to the MOGREPS short-range ensemble prediction system. Q. J. R. Meteorol. Soc. 135: 767–776
- MacLachlan, C., Arribas, A., et al.: Global Seasonal forecast system version 5 (GloSea5): a high-resolution seasonal forecast system, 2014, Q. J. Roy. Meteor. Soc., doi:10.1002/qj.2396
- Mogensen K, Balmaseda M, Weaver AT, Martin M, Vidard A. 2009. NEMOVAR: A variational data assimilation system for the NEMO ocean model. In ECMWF Newsletter, Walter Z. (ed.) 120: 17–21. ECMWF: Reading, UK.
- Mogensen K, Balmaseda MA, Weaver AT. 2012. ‘The NEMOVAR ocean data assimilation system as implemented in the ECMWF ocean analysis for System 4’, Technical Report TR-CMGC-12-30. CERFACS: Toulouse, France.
- Williams, K. D., Harris, C. M., Bodas-Salcedo, A., Camp, J., Comer, R. E., Copsey, D., Fereday, D., Graham, T., Hill, R., Hinton, T., Hyder, P., Ineson, S., Masato, G., Milton, S. F., Roberts, M. J., Rowell, D. P., Sanchez, C., Shelly, A., Sinha, B., Walters, D. N., West, A., Woollings, T., and Xavier, P. K.: The Met Office Global Coupled model 2.0 (GC2) configuration, Geosci. Model Dev., 8, 1509-1524, doi:10.5194/gmd-8-1509-2015, 2015.
...