On classifying rain types using satellite microwave observations

Varma, A. K., and G. Liu (2010), On classifying rain types using satellite microwave observations, J. Geophys. Res., 115, D07204, doi:10.1029/2009JD012058.
Abstract: 

Classification of rain type in satellite microwave observations is useful for various studies ranging from numerical weather prediction and precipitation climatology to satellite retrieval of rain amounts. In this study we have first examined the possibility of determining the distribution of convective/stratiform rain within a typical microwave radiometric pixel size area represented by the Tropical Rainfall Measuring Mission Microwave Imager (TMI) and then formulated an empirical relation between the convective to stratiform ratio and observed brightness temperatures. Rain classification with satellite microwave observation is hampered by the small size of the rain events. It is found from the rain observations during July 2000 that a significant number of 53% convective and 28% stratiform rain fill less than one fourth of the TMI pixel size area. The nonlinear relationship between brightness temperature and rain rate, along with horizontal and vertical inhomogeneity of the rain type distribution within the pixel, makes it difficult to work out the exact proportion of convective to stratiform distribution within the pixel. Here an algorithm is proposed to determine rain type on the basis of regression with 10 functions of 19, 37, and 85 GHz channels into three broad convective‐ stratiform proportions. This algorithm is able to identify rain types in about 70% of the TMI pixels accurately. To broaden the utility of the proposed method, a procedure has been developed by which the method can be applied to any other microwave radiometers with similar channels to TMI. Using this procedure, a successful application of the algorithm to Special Sensor Microwave Imager observations is demonstrated.

PDF of Publication: 
Download from publisher's website.