Extended Abstract (2.4M)P1.30
Imager capability on cloud classification using MODIS
Zhenglong Li, CIMSS/Univ. of Wisconsin, Madison, WI; and J. Li, P. Menzel, and T. J. Schmit
An automated cloud classification of surface and cloud types using radiance measurements with a maximum likelihood method is applied to study the capability of cloud classification by different sensors ----- the Advanced Baseline Imager (ABI), the Advanced Very High Resolution Radiometer/3 (AVHRR/3), Fengyun-1C (FY1C), Geostationary Operational Environmental Satellite-12 (GOES-12), Meteosat Second Generation (MSG1) and the Visible Infrared Imaging Radiometer Suite (VIIRS). The Moderate Resolution Imaging Spectroradiometer (MODIS) cloud mask is used as the training set (initial classification). Three different cases: a high latitude case, a hurricane case and a desert case, are studied. The results are compared with true color images, RGB(0.65, 1.6, 11um flipped) images, as well as GOES-8 Satellite Image Animation. Among all the six sensors, ABI usually shows the best capability since it has more spectral bands than the others. MSG1 and VIIRS have the similar capability on cloud classification, although they have some different bands. AVHRR and FY1C have more acceptable results than GOES because AVHRR and FY1C have both visible bands and 3.7/3.9um bands, which are of great importance for cloud classification.
Poster Session 1, IIPS Poster Session I
Monday, 10 January 2005, 2:30 PM-4:00 PM
Previous paper Next paper