|
The direct assimilation of satellite-measured infrared radiances into numerical weather prediction and cloud sounding applications is currently prohibited when these measurements include cloud radiative effects. The difficulty arises from the microphysical complexity of clouds and their radiative responses that are only now being adequately modeled for current and next generation satellite sensors. The parameterization of cloud properties to deliver much needed im-provements in speed and accuracy of forward radiative transfer models is still under development. Indirect use of cloud-contaminated radiances by way of cloud-cleared radiances has thus become the initial focus of efforts to improve the spatial density of useful satellite radiance measurements. This is particularly important for satellite sensors with relatively wide fields of view as the probability of entirely cloud-free observations can be surprisingly low. Two classes of cloud-cleared radiance retrieval approaches developed so far comprise the synergistic use of 1) collo-cated infrared and microwave measurements, and 2) collocated infrared imaging and sounding measurements. For example, AIRS/AMSU and AIRS/MODIS cloud-cleared algorithms are being demonstrated by NASA Earth Observing System and are to be adopted by NPP/NPOESS that have similar measurements available from the instrument suites CrIS/ATMS and CrIS/VIIRS. In this paper, the characteristics of these cloud-cleared radiances and their potential for numerical weather prediction and cloudy sounding applications are evaluated. Preliminary results have shown that these two approaches, though quite different in character, are both effective and complementary. Where microwave measurements are unavailable, the synergistic imaging/sounding approach to cloud-clearing is the only reliable indirect use of cloud-contaminated infrared measurements, as is the case for geostationary platforms due to the antenna requirements for a meteorologically useful microwave radiometer at 35,000 km.
|
