This paper describes the implementation of a method for adaptive estimation and correction of radiance biases in the ECMWF variational data assimilation system. Biases are partly caused by problems with the measurements themselves, but are also affected by errors in the radiative transfer calculations that are used to simulate radiance observations from the model state. These errors are different for each sensor and each channel, and tend to depend on the state of the instrument (e.g., scan position, poor calibration) as well as on local properties of the geophysical parameters being sensed (McNally et al., 2000). Over the years, sufficiently effective schemes have been developed for screening the data and estimating their biases, so that the quality-controlled and biascorrected radiance data can be usefully assimilated in an NWP system. At the same time, the number and variety of available sensors and the quantity of measurements they produce have increased to the point that the processing and management of the data now presents a tremendous challenge.
[1]
John Derber,et al.
The Use of TOVS Cloud-Cleared Radiances in the NCEP SSI Analysis System
,
1998
.
[2]
A. Mcnally,et al.
A cloud detection algorithm for high‐spectral‐resolution infrared sounders
,
2003
.
[3]
Graeme Kelly,et al.
A satellite radiance‐bias correction scheme for data assimilation
,
2001
.
[4]
M. Matricardi,et al.
An improved fast radiative transfer model for assimilation of satellite radiance observations
,
1999
.
[5]
John Derber,et al.
The use of TOVS level‐1b radiances in the NCEP SSI analysis system
,
2000
.
[6]
Jean-Noël Thépaut,et al.
Estimating atmospheric CO2 from advanced infrared satellite radiances within an operational 4D-Var data assimilation system: Methodology and first results
,
2004
.