METEOROLOGICAL DATA PERSPECTIVE
355
3.4
Channel selection for satellite sounders
Another optimisation of the use of observations can be the selection of
the most valuable subset of data, if some of the global observing systems are
providing too many pieces of information for the data processing
capabilities. In particular, advanced infrared sounders provide thousands of
radiance data at every observation location. The first instrument with kilochannel data is the Atmospheric InfraRed Sounder (AIRS) on the Aqua
satellite launched by the National Aeronautics and Space Administration
(NASA) in 2002. On the European side, the French space agency Centre
National d’Etudes Spatiales (CNES) and the European Meteorological
Satellite organization (EUMETSAT) have developed the Infrared
Atmospheric Sounding Interferometer (IASI) to be launched at the end 2005.
For operational NWP systems, these data will provide temperature and
humidity information with a fine vertical resolution. The number of
individual pieces of information is not usable in an operational NWP
context, and several possibilities are being investigated to choose an
“optimal” subset of data. This would allow to extract the maximum
information content from hyperspectral sounders, with a reduced number of
individual data. Solutions proposed to solve this problem include the
selection of relevant limited spectral bands (Aires et al, 2002), the grouping
of highly correlated channels in the same spectral area into super-channels,
the use of a partial eigen-decomposition of the radiance data (Joiner and Da
Silva, 1998), and the selection of individual channels based on objective
criteria (Rodgers, 1996).
Figure 7. Location of 300 channels (black lines) selected by an iterative procedure based on
information content for the retrieval of temperature and humidity for a IASI spectrum (red
line). From Rabier et al (2002).
An example of such a channel selection procedure is presented in Rabier
et al. (2002). The selection of individual channels is performed for simulated
355
3.4
Channel selection for satellite sounders
Another optimisation of the use of observations can be the selection of
the most valuable subset of data, if some of the global observing systems are
providing too many pieces of information for the data processing
capabilities. In particular, advanced infrared sounders provide thousands of
radiance data at every observation location. The first instrument with kilochannel data is the Atmospheric InfraRed Sounder (AIRS) on the Aqua
satellite launched by the National Aeronautics and Space Administration
(NASA) in 2002. On the European side, the French space agency Centre
National d’Etudes Spatiales (CNES) and the European Meteorological
Satellite organization (EUMETSAT) have developed the Infrared
Atmospheric Sounding Interferometer (IASI) to be launched at the end 2005.
For operational NWP systems, these data will provide temperature and
humidity information with a fine vertical resolution. The number of
individual pieces of information is not usable in an operational NWP
context, and several possibilities are being investigated to choose an
“optimal” subset of data. This would allow to extract the maximum
information content from hyperspectral sounders, with a reduced number of
individual data. Solutions proposed to solve this problem include the
selection of relevant limited spectral bands (Aires et al, 2002), the grouping
of highly correlated channels in the same spectral area into super-channels,
the use of a partial eigen-decomposition of the radiance data (Joiner and Da
Silva, 1998), and the selection of individual channels based on objective
criteria (Rodgers, 1996).
Figure 7. Location of 300 channels (black lines) selected by an iterative procedure based on
information content for the retrieval of temperature and humidity for a IASI spectrum (red
line). From Rabier et al (2002).
An example of such a channel selection procedure is presented in Rabier
et al. (2002). The selection of individual channels is performed for simulated