Premium
Adjoint sensitivity analysis of an observational operator for visible and infrared cloudy‐sky radiance assimilation
Author(s) -
Greenwald Thomas J.,
Vukićević Tomislava,
Grasso Louis D.,
Haar Thomas H. Vonder
Publication year - 2004
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1256/qj.03.44
Subject(s) - radiance , environmental science , radiative transfer , data assimilation , meteorology , numerical weather prediction , sky , remote sensing , wavelength , cloud fraction , ice cloud , satellite , infrared , atmospheric sciences , cloud cover , cloud computing , physics , geology , optics , astronomy , computer science , operating system
An adjoint modelling system is developed for an observational operator at visible and infrared wavelengths to explore the connection between cloud microphysics and top of atmosphere (TOA) radiances at cloud‐resolving scales (2–5 km) in preparation for direct assimilation of cloudy‐sky radiance satellite data. Analysis was performed on complex simulated three‐dimensional cloud fields for different weather phenomena generated by the Regional Atmospheric Modeling System using two‐moment microphysics. Sensitivity of TOA radiances at 0.63, 3.92, and 10.7 µm to changes in cloud mixing ratio revealed that small liquid drops and ice particles for very optically thin clouds were the largest contributors to the radiative sensitivities. More importantly, the sensitivities at these wavelengths were found to be complementary; i.e. 0.63 µm reflectances possessed greatest sensitivity to optically thinner water and ice clouds, whereas 3.92 µm responded to thick water clouds and to ice clouds, while 10.7 µm was most sensitive to thinner, cold ice clouds. Implications for numerical weather prediction (NWP) models that do not predict particle number concentration are that radiative sensitivities change somewhat in magnitude but retain the same sign, provided reasonable concentrations are assumed for broad classes of particle types. Overall, results indicated that satellite radiances measured in visible/infrared spectral windows contain potential information regarding cloud microphysics, especially at solar wavelengths, suggesting that direct assimilation of these data may be useful in supplying unique cloud information to NWP models. Copyright © 2004 Royal Meteorological Society