The Design of NCEP GEFS Reforecasts to Support Subseasonal and Hydrometeorological Applications

Regular generation of Reanalysis/Reforecast has been recognized as integral to the implementation of the Global Ensemble Forecast System (GEFS) at the NOAA/National Centers for Environmental Prediction (NCEP) Environmental Modeling Center. In the next GEFS implementation, a 30-year (1989-2018) reforecast will be generated to support forecast validation and calibration for various stakeholders, private sectors, and public users. The service areas will cover daily weather, week-2 forecasts, subseasonal forecasts, and hydrometeorological applications. However, only 19 years (2000-2018) of new reanalysis data are available due to limited computer resources and inconsistency of satellite observations within the past 30 years (i.e. AMUSU data was introduced into the regular observation system in 1999). Meanwhile, there are also some differences between the reanalyses and real-time analyses. Therefore, the impact of all these differences on model systematic error as a function of forecast lead-time will be investigated and posted for users.

The study will focus on near-surface temperatures (T2m) in three areas which are 1) systematic differences from three analyses (CFSR – Climate Forecast System Reanalysis; FV3 based reanalysis; FV3 based retrospective analysis) applied to the same period; 2) the qualities of ensemble reforecasts (skills and bias) that are impacted by the different initial analyses and perturbations within the same forecast model; 3) the characteristics of model systematic error, in terms of spatial distribution and forecast lead-time. Results will be documented as public references.