The First Global Climatology of Cloud Layering Derived from MODIS Using a New Algorithm

Fu-Lung Chang

ESSIC/UMD

Abstract:

Accurate representation of cloud vertical distribution is critical in global climate modeling. To date, large discrepancies still exist in our knowledge of the global cloud vertical structure. Satellite global cloud climatology has been generated mainly based on the data from the International Satellite Cloud Climatology Project (ISCCP), but the ISCCP data are limited by its conventional use of a single infrared channel to infer cloud top height. While the infrared method is useful for opaque clouds, it cannot resolve for non-opaque clouds like the cirrus.  The new millennial NASA Moderate-resolution Imaging Spectrometer (MODIS) employs a CO2-slicing technique to retrieve more accurate cloud-top height for cirrus cloud.  However, both operational MODIS and ISCCP algorithms assume that clouds were all single-layered.  Using a new retrieval algorithm developed by Chang and Li (2005), a new MODIS-based cloud product is generated.  It can deal with cirrus overlapping low cloud conditions and retrieve individual optical properties for the cirrus and low clouds.  The main objectives of our study are to examine the frequency occurrences of high, mid, and low clouds and assess the differences in the global cloud climatologies that result from the ISCCP, MODIS, and GCMs. Significant differences are revealed between the satellite products and the GCM results. Our new algorithm can be modified for operational applications to the NPOESS and GOES-R data.



.

.