Cirrus clouds cover 30% of the globe at any given time. These clouds play an important role in regulating climate, e.g. the radiative properties and extent of thin cirrus clouds associated with regions of tropical convection heavily influence the outcomes of the much- debated global warming scenarios. This paper introduces an optimal estimation based retrieval for locating thin cirrus and determining cloud optical depth, effective radius, and temperature from satellite data. The retrieval is based on the split window technique where effective radius and optical depth are retrieved from observed 10.8 and 12.0 micron brightness temperatures. Traditionally, large errors have resulted in this approach from a lack of knowledge of cloud emission temperature. Within the optimal estimation framework, however, it is possible to use explicit cloud boundary information to constrain the retrieval, significantly reducing the error in the estimates of cloud optical depth and effective radius.

To highlight potential applications of our retrieval, TRMM data was used to develop estimates of cirrus cloud extent and microphysical properties over the southern hemisphere tropics. These simple climatologies were then combined with with TRMM TMI-based instantaneous rainfall product (2A12) to explore the relationship between convection and cirrus clouds on a variety of spatial and temporal scales. The estimates of cloud temperature from passive TRMM measurements used to create these climatologies, however, are relatively inaccurate. With the launch of a cloud radar on CloudSat in spring 2004, a much more accurate means to estimate cloud boundary information will become available for most of the Southern Hemisphere. Co-located cloud radar observations from the CloudSat satellite and infrared radiances from the MODIS instrument aboard the Earth Observing System (EOS) Aqua satellite will allow the retrieval presented here to be implemented on a global scale. An advanced version of this retrieval could be combined with on-going work at Colorado State University to develop a global retrieval of the vertical distribution of cirrus cloud properties. A more accurate estimate of the horizontal and vertical distribution of cirrus cloud properties would be of obvious use in many climatological studies of Southern Hemisphere atmospheric processes.