The ECMWF Re-Analysis (ERA) project
ECMWF has provided global atmospheric analyses from its archive for many
years. The ERA project was devised in response to wishes expressed by many users
for a data set generated by a modern, consistent, and invariant data
assimilation system.
ERA has received assistance from the ECMWF Council, the European Community,
the University of California, the Japan Meteorological Agency (JMA), Cray
Research Incorporated, and the Center for Ocean-Land-Atmosphere Studies
(COLA).
In addition data and co-operation have been forthcoming from many quarters,
including ECMWF Member States, the University of Innsbruck, the National
Ocean and Atmospheric Administration (NOAA) and National Center for Atmospheric
Research (NCAR) in the United States of America, and the Australian Bureau
of Meteorology (BoM).
Data sources
The real time data WMO World Weather Watch data from ECMWF's archive (which
has a 3 day cut-off) are augmented by the following additional data:
- NOAA satellite radiance data for the 15 year period.
- The Comprehensive Ocean-Atmosphere Data Set (COADS) ship data.
- FGGE and ALPEX data.
- GMS cloud winds, AIREP and TEMP data supplied by JMA.
- Pseudo observations data from BoM.
Sea surface temperature analyses for ERA are provided by the United
Kingdom Meteorological Office for the early period, and by NOAA from
November 1981 onwards. Sea ice cover has been derived from SMMR and SSM/I
satellite data.
The Assimilation system
The ERA data assimilation system is a special version of the ECMWF
operational data assimilation system which includes:
- Intermittent statistical (Optimum Interpolation) analysis with 6 hour
cycling.
- One dimensional variational (1D-VAR) physical retrieval of TOVS cloud
cleared radiance (CCR) data.
- Diabatic, non-linear normal mode initialisation (five vertical
modes).
- A spectral T106 forecast model with 31 hybrid vertical levels, and a
fully three dimensional semi-Lagrangean advection scheme.
- A physical parametrization package which includes:
- mean orography with a compatible parametrization of sub-grid scale orography;
- a four layer prognostic soil scheme with no external
forcing;
- an interactive cloud/radiation scheme with full model
representation of cloud water content and cloud cover;
- a planetary boundary layer parametrization based on
similarity.