CLOUD PROPERTIES INFERRED FROM 8-12-MU-M DATA

A trispectral combination of observations at 8-, 11-, and 12-mum bands is suggested for detecting cloud and cloud properties in the infrared . Atmospheric ice and water vapor absorption peak in opposite halves o f the window region so that positive 8-minus-11-mum brightness tempera ture differences indicate cloud, while near-zero or negative differenc es indicate clear regions. The absorption coefficient for water increa ses more between 11 and 12 mum than between 8 and 11 mum, while for ic e, the reverse is true. Cloud phase is determined by a scatter diagram of 8-minus-11-mum versus 11-minus-12-mum brightness temperature diffe rences; ice cloud shows a slope greater than 1 and water cloud less th an 1. The trispectral brightness temperature method was tested upon hi gh-resolution interferometer data resulting in clear-cloud and cloud-p hase delineation. Simulations using differing 8-mum bandwidths reveale d no significant degradation of cloud property detection. Thus, the 8- mum bandwidth for future satellites can be selected based on the requi rements of other applications, such as surface characterization studie s. Application of the technique to current polar-orbiting High-Resolut ion Infrared Sounder (HIRS)-Advanced Very High Resolution Radiometer ( AVHRR) datasets is constrained by the nonuniformity of the cloud scene s sensed within the large HIRS field of view. Analysis of MAS (MODIS A irborne Simulator) high-spatial resolution (500 m) data with all three 8-, 11-, and 12-mum bands revealed sharp delineation of differing clo ud and background scenes, from which a simple automated threshold tech nique was developed. Cloud phase, clear-sky, and qualitative differenc es in cloud emissivity and cloud height were identified on a case stud y segment from 24 November 1991, consistent with the scene. More rigor ous techniques would allow further cloud parameter clarification. The opportunities for global cloud delineation with the Moderate-Resolutio n Imaging Spectrometer (MODIS) appear excellent. The spectral selectio n, the spatial resolution, and the global coverage are all well suited for significant advances.