1. Introduction / History
The Global Ensemble Forecast System (GEFS) has been run in NCEP's Production Suite since December 1992, using the NCEP Global Spectral Model and initially using the Breeding Vector (BV) technique to generate perturbations in the initial conditions. GEFS runs four times per day (0000, 0600, 1200 and 1800 UTC) out to 16 days. At each time, 10 (5 pairs) perturbed members are initialized using BV method, cycling every 6 hours. If tropical cyclones are present in the initial conditions, a tropical storm relocation technique is applied to each ensemble member to adjust the initial central location to the observed location (see: Liu and et al., 2006). An extended BV method with Ensemble Transform and Rescaling (BV-ETR; Wei and et al., 2008) was implemented operationally in 2006. In early 2010, GEFS included model uncertainty using the Stochastic Total Tendency Perturbation (STTP) algorithm (see: Hou and et al., 2012). Starting in December 2015, GEFS initial condition perturbations were selected from the operational hybrid Global Data Assimilation System (GDAS) 80-member Ensemble Kalman Filter (EnKF; Whitaker et al., 2008) 6-h forecast, which includes tropical storm relocation and centralization of the initial perturbations (Zhao et al., 2016).
2. Recent Changes in Configuration (Feb. 2012 and Dec. 2015)
Users can refer to NCEP Global Ensemble Forecast system implementation log for details.
3. Horizontal and Vertical Resolution:
In the December 2015 implementation, the horizontal resolution has increased to TL574 (about 34km on the equator) for 0-192 hours and TL372 (about 55km) for 192-384 hours. The vertical resolution has also increased from 42 to 64 hybrid levels for all forecast hours.
There are 20 perturbed members (plus the ensemble control member) for all four forecast cycles.
5. Generation of the Initial Perturbations:
Initial perturbations are generated from the EnKF component of NCEP Global Data Assimilation System. GEFS uses the 6-h 80-member ensemble forecast instead of the EnKF analysis due to timing constraints within the NCEP production suite (Zhou et al., 2016).
6. Representation of Model Related Uncertainty:
In Feb. 2010, a Stochastic Total Tendency Perturbation scheme was implemented to represent uncertainties associated with the forecast model. STTP is based on the hypothesis that tendencies of the ensemble perturbations provide a representative sample of the random total model errors (see: Hou et al. 2012). In the latest Dec. 2015 upgrade, an additional tuning process is applied for lower latitudes (tropical area) and no perturbation is applied for tropical surface pressure (Zhou et al., 2016).
7. Where to find GEFS data
Gridded GEFS data is available through the NOAA National Operational Model Archive and Distribution System (NOMADS). NOMADS also contributes GEFS ensemble data to the The Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble (TIGGE) by calculating a dozen World Meteorological Organization (WMO)-required variables and passing them to the National Center for Atmospheric Research (NCAR) for permanent archive. NOMADS also provides an additional tool, the NOMADS Ensemble Probability Tool, which allows a user to query the multiple forecast ensemble to determine the probability that a set of conditions will occur at a given location using all of the GEFS ensemble members in near-real time.
For a detailed list of GEFS output available on NOMADS and the file inventories, see the NCO GEFS Product Inventory Page
Other NCEP GEFS web pages:
- Official GEFS Graphics at the NCEP Model Analysis and Guidance Page : Mean/Spread and Spaghetti Plots
- Experimental GEFS plume diagram web page
- GEFS FAQ Page
- See the "Comprehensive List of Ensemble Web Sites" link above for a list of US (Government/Academic) and international ensemble information
- GEFS Verification Sites:
- NCEP Global Model Verification Portal, including GEFS and NAEFS Verification Pages
- NAEFS Verification Statistics (updated every quarter) ; Include, NCEP/GEFS, CMC/GEFS, FNMOC/GEFS, ECMWF/GEFS, raw and bias corrected forecast for 00UTC initial forecast each day
- NUOPC verification statistics (updated seasonally)
- Daily Comparisons between Operational Global Model Forecasts : 00Z NCEP, 12Z ECMWF, 00Z UKMet, 00Z MSC, 00Z NOGAPS
- Daily Comparisons between Operational NCEP Global Model & Experimental Runs
- Daily Comparisons between GFS/GEFS control & ECMWF/ECMWF control ; 00Z T382/38km GFS, 00Z T190/70km GEFS control
12Z T1279/16km ECMWF, 12Z T639/30km ECMWF ensemble control
- Daily Values of 500 hPa Height AC, RMS, Talagrand & Outliers : Mean of 14 GFS, 10 ECMWF and 16 CMC members
GFS operational, GEFS control
- Long-term Forecast Skill Comparisons of 500 hPa Height & 850 hPa Temperature AC and CRPSS : NCEP vs CMC vs ECMWF
00Z, NH only
- Comparison between Raw and Calibrated GFS QPF/PQPF Forecasts
- Comparison between NCEP and CMC QPF/PQPF Forecasts
- Comparison between NCEP and CMC PQPF Forecast Scores
Please send any comments and suggestions about GEFS to Yuejian Zhu.
References for GEFS, NAEFS and post processing:
Link to a comprehensive list of publications since 1995
Cui, B., Z. Toth, Y. Zhu and D. Hou, 2012: Bias Correction For Global Ensemble Forecast, Weather and Forecasting, 27, 396-410
Cui, B., Y. Zhu, Z. Toth and D. Hou, 2018: Development of Statistical Post-processor for NAEFS. To be submitted to Weather and Forecasting
Guan, H., B. Cui, Y. Zhu, 2015: Improvement of Statistical Postprocessing Using GEFS Reforecast Information, Weather and Forecasting, Vol. 30, 841-854
Hou, D., Z. Toth, Y. Zhu, W. Yang and R. Wobus, 2012: "A Stochastic Total Tendency Perturbation Scheme Representing Model- Related Uncertainties in the NCEP Global Ensemble Forecast System" Submitted to Tellus-A)
Hou, D., M. Charles, Y. Luo, Z. Toth, Y. Zhu, R. Krzysztofowicz, Y. Lin, P. Xie, D-J. Seo, M. Pena and B. Cui, 2012: Climatology-Calibrated Precipitation Analysis at Fine Scales: Statistical Adjustment of STAGE IV towards CPC Gauge-Based Analysis, Jaurnal of Hydrometeorology Vol. 15 2542-2557
Liu, Q., S. J. Lord, N. Surgi, Y. Zhu, R. Wobus, Z. Toth and T. Marchok, 2006: Hurricane Relocation in Global Ensemble Forecast System, Preprints, 27th Conf. on Hurricanes and Tropical Meteorology, Monterey, CA, Amer. Meteor. Soc., P5.13.
Ma, J., Y. Zhu, D. Wobus and P. Wang, 2012: An Effective Configuration of Ensemble Size and Horizontal Resolution for the NCEP GEFS, Advance in Atmospheric Sciences, Vol. 29, No. 4, 782-794
Ma, J., Y. Zhu, D. Hou, X. Zhou and M. Pena, 2014: Ensemble Transform with 3D Rescaling Initialization Method, Monthly Weather Review, Vol. 142, 4053-4073
Palmer, T. N., R. Buizza, F. Doblas-Reyes, T. Jung, M. Leutbecher, G. Shutts, M. Steinheimer, and A. Weisheimer, 2009: Stochastic Parametrization and Model Uncertainty. ECMWF Tech. Memo. 598, 44.
Shutts, G., 2005: A kinetic energy backscatter algorithm for use in ensemble prediction systems. Quart. J. Roy. Meteor. Soc., 131, 3079-3102.
Toth, Z., and E. Kalnay, 1997: Ensemble forecasting at NCEP and the breeding method. Mon. Wea. Rev., 125, 3297-3319.
Wei, M., Z. Toth, R. Wobus, and Y. Zhu, 2008: Initial Perturbations Based on the Ensemble Transform (ET) Technique in the NCEP Global Operational Forecast System, Tellus 59A, 62-79
Whitaker, Jeffrey S., Thomas M. Hamill, Xue Wei, Yucheng Song, Zoltan Toth, 2008: Ensemble Data Assimilation with the NCEP Global Forecast System. Mon. Wea. Rev., 136, 463-482.
Zhou, X., Y. Zhu, D. Hou, Y. Luo, J. Peng and R. Wobus, 2017: The NCEP Global Ensemble Forecast System with the EnKF Initialization. Wea. Forecasting, 32, 1989-2004.
Zhu Y., and Y. Luo, 2014: Precipitation Calibration Based on Frequency Matching Method (FMM), Weather and Forecasting, Vol. 30, 1109-1124
National Weather Service Web Page Disclaimer